IN THE UNITED STATES PATENT AND TRADEMARK OFFICE
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`
`ARGENTUM PHARMACEUTICALS LLC,
`Petitioner,
`
`v.
`
`RESEARCH CORPORATION TECHNOLOGIES, INC.,
`Patent Owner.
`
`Case No. IPR2016-00204
`Patent No. RE 38,551
`
`PATENT OWNER RESPONSE
`PURSUANT TO 37 C.F.R. § 42.120
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`
`
`I.
`
`II.
`
`III.
`
`TABLE OF CONTENTS
`
`Page(s)
`
`Introduction ..................................................................................................... 1
`
`The Development of the Inventions and the ’551 Patent ............................... 6
`
`Petitioner Fails to Show that Any Claim of the ’551 Patent Is
`Unpatentable Over Kohn 1991, Silverman, and the ’729 Patent ................... 9
`
`A.
`
`B.
`
`Petitioner fails to show that a POSA would have selected a
`functionalized amino acid (FAA) as a lead compound. ....................... 9
`
`Petitioner fails to show that a POSA would have selected
`Compound 3l as a lead compound. .................................................... 14
`
`1.
`
`2.
`
`3.
`
`The FAA prior art taught that heteroaromatic FAAs were
`preferred. .................................................................................. 15
`
`A POSA would not have selected a compound containing
`a nitrogen-oxygen bond as a lead compound. ......................... 17
`
`Petitioner and Dr. Wang rely on impermissible hindsight
`to select Compound 3l as a lead compound. ............................ 19
`
`C.
`
`Petitioner fails to show that a POSA, even if looking to FAAs
`and selecting Compound 3l, would have modified Compound 3l
`into any of the compounds of claims 1–9, and would have done
`so with a reasonable expectation of success. ..................................... 23
`
`1.
`
`2.
`
`3.
`
`A POSA would not have been motivated to modify the
`amino moiety at the α-carbon of Compound 3l. ...................... 24
`
`A POSA modifying Compound 3l would have had no
`motivation to use bioisosterism at the α-carbon. ..................... 27
`
`A POSA would not have had any reasonable expectation
`of success in employing bioisosterism to modify
`Compound 3l. ........................................................................... 33
`
`Page i
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`a)
`
`b)
`
`Petitioner and Dr. Wang’s analysis relies on
`hindsight and fails to acknowledge the
`unpredictability of bioisosterism. .................................. 33
`
`The evidence shows that a POSA would not have
`reasonably expected success in substituting
`methoxyamino with methoxymethyl. ............................ 36
`
`Petitioner relies on information not available in the prior
`art to assert motivation and reasonable expectation of
`success in modifying Compound 3l. ........................................ 42
`
`A POSA would not have been motivated to isolate the R-
`isomer. ...................................................................................... 43
`
`4.
`
`5.
`
`D.
`
`E.
`
`Petitioner fails to show that any prior art discloses the
`therapeutic composition of Claim 10, and that a POSA would
`have achieved the therapeutic composition of Claim 10 with a
`reasonable expectation of success. ..................................................... 45
`
`Petitioner fails to show that a POSA would have achieved the
`methods of treatment of Claims 11–13 with a reasonable
`expectation of success. ....................................................................... 49
`
`B.
`
`IV. Objective Indicia Confirm that the Inventions of Claims 1-13 Are Not
`Obvious ......................................................................................................... 51
`A. Vimpat®’s unexpected results have received extensive industry
`praise................................................................................................... 52
`Vimpat®’s development was marked by skepticism and failure
`of others. ............................................................................................. 54
`Vimpat® met a long-felt need. ............................................................ 56
`C.
`D. Vimpat® is a commercial success. ...................................................... 58
`E.
`Sixteen generic drug companies have sought to market copies
`of Vimpat®. ......................................................................................... 60
`Conclusion .................................................................................................... 61
`
`V.
`
`
`
`Page ii
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`TABLE OF AUTHORITIES
`
`Cases
`Alcon Research, Ltd. v. Apotex Inc.,
`687 F.3d 1362 (Fed. Cir. 2012) .......................................................................... 58
`
`Page(s)
`
`Crocs, Inc. v. Int’l Trade Comm’n,
`598 F.3d 1294 (Fed. Cir. 2010) .......................................................................... 53
`
`In re Cyclobenzaprine Hydrochloride Extended-Release Capsule
`Patent Litig.,
`676 F.3d 1063 (Fed. Cir. 2012) .......................................................................... 51
`
`Daiichi Sankyo Co. v. Matrix Labs., Ltd.,
`619 F.3d 1346 (Fed. Cir. 2010) .................................................................... 10, 21
`
`In re Dow Chem. Co.,
`837 F.2d 469 (Fed. Cir. 1988) ............................................................................ 55
`
`Eisai Co. v. Dr. Reddy’s Labs., Ltd.,
`533 F.3d 1353 (Fed. Cir. 2008) ............................................................................ 9
`
`Innopharma Licensing, Inc. v. Senju Pharm. Co., Ltd.,
`IPR2015-00902, Paper 90 (PTAB July 28, 2016) ........................................ 51, 58
`
`Intelligent Bio-Systems, Inc. v. Illumina Cambridge Ltd.,
`821 F.3d 1359 (Fed. Cir. 2016) ...................................................................... 5, 47
`
`J.T. Eaton & Co. v. Atl. Paste & Glue Co.,
`106 F.3d 1563 (Fed. Cir. 1997) .......................................................................... 58
`
`In re Magnum Oil Tools Int’l, Ltd.,
`__ F.3d __, 2016 WL 3974202 (Fed. Cir. 2016) .................................................. 1
`
`Mintz v. Dietz & Watson, Inc.,
`679 F.3d 1372 (Fed. Cir. 2012) .......................................................................... 51
`
`Otsuka Pharm. Co. v. Sandoz, Inc.,
`678 F.3d 1280 (Fed. Cir. 2012) ...................................................................passim
`
`Page iii
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`Pressure Prods. Med. Supplies, Inc. v. Greatbatch Ltd.,
`599 F.3d 1308 (Fed. Cir. 2010) .......................................................................... 54
`
`Procter & Gamble Co. v. Teva Pharm. USA, Inc.,
`566 F.3d 989 (Fed. Cir. 2009) ............................................................................ 56
`
`In re Rosuvastatin Calcium Patent Litig.,
`703 F.3d 511 (Fed. Cir. 2012) ............................................................................ 54
`
`In re Soni,
`54 F.3d 746 (Fed. Cir. 1995) .............................................................................. 52
`
`Spectrum Pharms. Inc. v. Sandoz Inc.,
`802 F.3d 1326 (Fed. Cir. 2015) .......................................................................... 44
`
`WBIP, LLC v. Kohler Co.,
`__F.3d __ (Fed. Cir. 2016), 2016 WL 3902668 ........................................... 51, 59
`
`Windsurfing Int’l, Inc. v. AMF, Inc.,
`782 F.2d 995 (Fed. Cir. 1986) ............................................................................ 60
`
`In re Youngblood,
`No. 98-1518, 1999 WL 504243 (Fed. Cir. 1999) ............................................... 57
`
`
`
`Page iv
`
`

`
`IPR2016-00204
`
`I.
`
`Introduction
`
`029819.0100-US03
`
`Patent Owner Research Corporation Technologies, Inc. (“Patent Owner”)
`
`provides the following response to the petition filed by Argentum Pharmaceuticals
`
`LLC (“Petitioner”) on November 23, 2015, requesting inter partes review of
`
`claims 1–13 of U.S. Reissued Patent No. RE38,551 (“the ’551 patent”).1 On May
`
`23, 2016, the Board instituted inter partes review of claims 1–13 on the sole
`
`ground of obviousness over Kohn 1991 (Ex. 1012) Compound 3l, in light of
`
`Silverman (Ex. 1013) and the ’729 patent (Ex. 1009). As explained in this
`
`response, Petitioner fails to meet its burden of proving unpatentability of any claim
`
`of the ’551 patent. See In re Magnum Oil Tools Int’l, Ltd., __ F.3d __, 2016 WL
`
`3974202, at *6 (Fed. Cir. 2016) (“‘In an inter partes review, the burden of
`
`persuasion is on the petitioner to prove ‘unpatentability by a preponderance of the
`
`evidence,’ 35 U.S.C. § 316(e), and that burden never shifts to the patentee.’”)
`
`(internal citation omitted).
`
`
`
`
`
`
`1 Patent Owner Research Corporation Technologies licenses the ’551 patent to
`
`Harris FRC Corporation, who in turn licenses the ’551 patent to UCB, Inc.
`
`(“UCB”). UCB markets the VIMPAT® drug product, a commercial embodiment of
`
`the invention claimed in the ’551 patent.
`
`1
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`Petitioner has not shown that claims 1–9 of the ’551 patent are unpatentable
`
`over Kohn 1991 Compound 3l, in light of Silverman. Petitioner’s entire
`
`obviousness challenge is grounded on a single prior art data point—a comparison
`
`of Kohn 1991 Compounds 2a and 3a—that bears no relationship to either
`
`Compound 3l or the claimed compounds, including the claimed lacosamide
`
`compound. In an attempt to create a link between this sole prior art data point and
`
`lacosamide, Petitioner points to pharmacological data for lacosamide itself, which
`
`does not appear anywhere in the prior art. In addition, Petitioner’s bare assertions
`
`regarding bioisosterism and a POSA’s “expectations” are without support and
`
`ignore the accepted reality that bioisosterism—just one of many research tools
`
`available in 1996—was an unpredictable modification method. Petitioner’s efforts
`
`to craft an obviousness challenge out of 1) one unrelated prior art data point, 2)
`
`pharmacological data taken from the ’551 patent itself, and 3) unsupported and
`
`inaccurate statements regarding bioisosterism are based on hindsight and fall far
`
`short of establishing unpatentability of any of claims 1–9.
`
`Petitioner also fails to show that a POSA would have chosen a
`
`functionalized amino acid (“FAA”) as a lead compound in the first place. In
`
`choosing a lead compound, a POSA would have chosen a compound that was
`
`already FDA-approved for the treatment of epilepsy, or that was at least in
`
`advanced clinical development. But no FAA had been approved, and no FAA was
`
`2
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`in advanced clinical development. Instead, in 1996 FAAs were a class of
`
`compounds, exclusively pursued by a small research group headed by Dr. Kohn,
`
`for which only very limited data existed and for which no quantitative structure-
`
`activity relationship (“QSAR”) had been established. A POSA would not have
`
`selected a compound such as an FAA as a lead compound.
`
`Moreover, even if a POSA would have selected an FAA as a lead, Petitioner
`
`fails to show that a POSA would have selected one particular FAA—Compound
`
`3l—as a lead compound. First, Compound 3l had been publicly disclosed for
`
`nearly five years as of 1996, but had received no interest from the scientific
`
`community. Second, despite the compound’s apparent efficacy, a POSA would
`
`have recognized that the entirety of the FAA prior art pointed in a different
`
`direction—to structurally different heteroaromatic compounds, which, due to their
`
`high efficacy and low neurotoxicity, a POSA would have viewed as the most
`
`promising FAAs. Furthermore, notwithstanding Compound 3l’s activity, a POSA
`
`would have understood that the compound was undesirable as a pharmacological
`
`compound because its nitrogen-oxygen bond would be susceptible to breaking at
`
`physiological pH, rendering it an undesirable lead compound for a POSA looking
`
`to develop a new antiepileptic drug (“AED”). In fact, multiple FAA compounds
`
`exhibited similar or even better activity compared to Compound 3l but did not
`
`contain this problematic nitrogen-oxygen bond. The only explanation for
`
`3
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`Petitioner’s selection of Compound 3l over these other FAAs is Compound 3l’s
`
`structural similarity
`
`to
`
`the claimed compound
`
`lacosamide, which
`
`is an
`
`impermissible use of hindsight.
`
`In addition, a POSA would not have modified Compound 3l, or have had
`
`any reasonable expectation of success in modifying Compound 3l to achieve the
`
`claimed compounds, including the claimed lacosamide compound. A POSA would
`
`have recognized that the amino moiety (-NH-) at the α-carbon of Compound 3l
`
`was the source of the compound’s good anticonvulsant activity and would not have
`
`sought to replace it, especially with a methylene moiety (-CH2-), which the FAA
`
`prior art uniformly demonstrated would adversely affect activity. A POSA would
`
`have also recognized that the outcomes of “bioisosteric” replacements were highly
`
`unpredictable.
`
`Indeed, a real-life POSA, Eli Lilly, rejected both Compound 3l and
`
`Petitioner’s proposed modification to it. In a brief collaboration involving Dr.
`
`Kohn and Eli Lilly, Dr. Kohn sent lists of FAA compounds to Eli Lilly for
`
`potential development, and those FAA compounds included Compound 3l and
`
`Petitioner’s “racemic lacosamide.” See Pet. at 46. But Eli Lilly declined to pursue
`
`either compound.
`
`With respect to claim 10 of the ’551 patent, Petitioner faces a complete
`
`failure of proof to show unpatentability over Kohn 1991 Compound 3l, in light of
`
`4
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`Silverman and the ’729 patent. The Board construed the term “therapeutic
`
`composition” in claim 10 as a composition “suitable for use as a treatment regimen
`
`over an extended period of time (chronic administration).” Paper 19 at 7–8.
`
`Although Petitioner was fully aware of this construction and in fact argued against
`
`it (Pet. at 7–10), Petitioner fails to offer any evidence—as there is none—that the
`
`combination of Kohn 1991, Silverman, and the ’729 patent provide such a
`
`“therapeutic composition.” Moreover, Petitioner’s failure of proof on this point is
`
`irremediable because Petitioner was obligated to “make [its] case” in the Petition
`
`and is therefore out of time to cure its lack of evidence. Intelligent Bio-Systems,
`
`Inc. v. Illumina Cambridge Ltd., 821 F.3d 1359, 1369 (Fed. Cir. 2016).
`
`Petitioner similarly fails to show that claims 11–13 of the ’551 patent are
`
`unpatentable over Kohn 1991 Compound 3l, in light of Silverman and the ’729
`
`patent. Claims 11–13 describe methods of treatment using the compounds of
`
`claims 1–9. Petitioner does not even attempt to show why a POSA would have
`
`been motivated to modify Compound 3l into the claimed compounds, including the
`
`claimed lacosamide compound—an FAA compound having no pharmacological
`
`data whatsoever in the prior art—with any reasonable expectation of success of
`
`achieving viable methods of treatment.
`
`Petitioner further fails to account for the fact that no FAA had ever been
`
`FDA-approved as an AED, and thus, there was a lack of clinical data
`
`5
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`demonstrating actual efficacy or acceptable toxicity for any FAA in a human. A
`
`POSA would have appreciated that efficacy and toxicity of a new class of AEDs in
`
`the patient population could not be predicted on the basis of the available data, and
`
`therefore a POSA would not have chosen an FAA for development in the first
`
`instance, or have had any reasonable expectation that such a compound would be
`
`suitable as a therapeutic composition or method of treatment in humans.
`
`In sum, Petitioner’s hindsight-driven arguments fail, and the patentability of
`
`the claims of the ’551 patent should be confirmed.
`
`II. The Development of the Inventions and the ’551 Patent
`
`The ’551 patent claims the compound lacosamide ((R)-N-benzyl-2-
`
`acetamido-3-methoxypropionamide, referred to in the ’551 patent as “BAMP”; Ex.
`
`1001, 24:56–58), therapeutic compositions comprising lacosamide, and methods of
`
`treating central nervous system (“CNS”) disorders, including epilepsy, by
`
`administering lacosamide. The ’551 patent provides the first disclosure of
`
`lacosamide, its strong anticonvulsant activity, low neurotoxicity, and in particular
`
`its low liver toxicity, which makes it ideal for the long-term administration
`
`required in the treatment of epilepsy.
`
`Epilepsy is an extremely heterogeneous disorder, with the particular needs of
`
`individual patients varying considerably, thus making the development of an
`
`effective and safe AED particularly challenging. See Ex. 2038 ¶¶20–25, 46–48. In
`
`6
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`addition to the difficulties of developing an effective and safe treatment for an
`
`extremely heterogeneous patient population, it is also important that potential
`
`AEDs are non-toxic during long-term treatment, because people with epilepsy
`
`usually need to take anticonvulsant drugs for decades. Id. at ¶48. As the ’551
`
`patent explains, an antiepileptic drug should ideally do more than provide high
`
`anticonvulsant activity, minimal neurological toxicity, and a high margin of safety
`
`after acute administration. Ex. 1001, 3:14–55. Long-term toxicity is also a critical
`
`consideration. Id. at 3:36–38.
`
`The historic failures to find and develop sufficiently safe and effective
`
`antiepileptic drugs prompted the National Institutes of Health (“NIH”) in 1975 to
`
`establish the Anticonvulsant Screening Program (“ASP”)2 to facilitate and
`
`encourage the discovery of new anticonvulsant agents. See, e.g., Exs. 2076, 2058,
`
`2075, 2130; Ex. 2036 ¶310. Yet, as of 1996, the priority date of the ’551 patent,
`
`only one of the 16,000 compounds screened—felbamate—had gained FDA
`
`approval. Exs. 2116, 2130; Ex. 2036 ¶311. But that product was later recognized to
`
`have “restricted value because of hematologic and hepatic toxicity.” Ex. 2001 at
`
`S21–S22. Just one year after felbamate’s market entry in 1993, the FDA required
`
`
`2 Currently known as the Epilepsy Therapy Screening Program (“ETSP”). See Ex.
`
`2130 (http://www.ninds.nih.gov/research/asp/index.htm).
`
`7
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`the labeling for felbamate to include a “black box” warning that it “should only be
`
`used in patients whose epilepsy is so severe” that the administering physician
`
`concluded that the risk of liver failure and aplastic anemia was acceptable. See Ex.
`
`2003.
`
`Dr. Harold Kohn, the inventor of the ’551 patent, conceived a new approach
`
`that was outside the mainstream of antiepileptic drug discovery. In the early 1980s,
`
`he theorized that a class of modified amino acids that he called “functionalized
`
`amino acids,” or “FAAs,” may demonstrate anticonvulsant activity, even though
`
`no known antiepileptic drug had this structure. See infra Section III.A; Ex. 2004 at
`
`568. When Dr. Kohn started his research, he had no evidence that any FAA would
`
`exhibit anticonvulsant activity, low or no neurological toxicity, a high margin of
`
`safety, and minimal adverse effects, such as low liver toxicity, during long-term
`
`chronic administration.
`
`In Dr. Kohn’s search for a new AED, he synthesized hundreds of FAA
`
`compounds. In the late 1980s he began collaborating with Eli Lilly to develop the
`
`compounds. Presented with lists of FAAs, including Compound 3l and an FAA
`
`with a methoxymethyl substituent at the α-carbon (the compound Petitioner refers
`
`to as “racemic lacosamide”), Eli Lilly chose an FAA with a heteroaromatic
`
`substituent at the α-carbon for development. However by 1991, Eli Lilly
`
`terminated the project because of its concerns over “sever[e] toxicity.” Ex. 2125.
`
`8
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`Dr. Kohn nonetheless persisted in his work.
`
`Dr. Kohn’s years of research culminated in the claimed inventions of the
`
`’551 patent, which represent significant advances in epilepsy treatment. The
`
`lacosamide invention met a long-recognized need to find a sufficiently safe and
`
`effective anticonvulsant drug for the long-term treatment of millions of patients
`
`afflicted with epilepsy for whom no available antiepileptic drugs could effectively
`
`treat their conditions. See, e.g., Exs. 2001, 2002; Section IV.C, infra. Indeed,
`
`following its introduction to the market, lacosamide has been praised as having
`
`“many favorable attributes, which may make it an optimal antiepileptic therapy.”
`
`Ex. 2102 at 13.
`
`III. Petitioner Fails to Show that Any Claim of the ’551 Patent Is
`Unpatentable Over Kohn 1991, Silverman, and the ’729 Patent
`A.
`
`Petitioner fails to show that a POSA would have selected a
`functionalized amino acid (FAA) as a lead compound.
`
`The “reasoned identification of a lead compound” is the first step in an
`
`obviousness analysis involving chemical compounds. Eisai Co. v. Dr. Reddy's
`
`Labs., Ltd., 533 F.3d 1353, 1359 (Fed. Cir. 2008). A lead compound “is a
`
`compound in the prior art that would be most promising to modify in order to
`
`improve upon its . . . activity and obtain a compound with better activity. . . . [A]
`
`lead compound is a natural choice for further development efforts.” Otsuka Pharm.
`
`Co. v. Sandoz, Inc., 678 F.3d 1280, 1291 (Fed. Cir. 2012) (internal quotation marks
`
`9
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`omitted). “In determining whether a chemist would have selected a prior art
`
`compound as a lead, the analysis is guided by evidence of the compound’s
`
`pertinent properties. Such properties may include positive attributes such as
`
`activity and potency, adverse effects such as toxicity, and other relevant
`
`characteristics in evidence.” Id. (internal citations omitted). “Absent a reason or
`
`motivation based on such prior art evidence, mere structural similarity between a
`
`prior art compound and the claimed compound does not inform the lead compound
`
`selection.” Id. Hindsight bias is improper: “the attribution of a compound as a lead
`
`compound after the fact must avoid hindsight bias; it must look at the state of the
`
`art at the time the invention was made to find a motivation to select and then
`
`modify a lead compound to arrive at the claimed invention.” Daiichi Sankyo Co. v.
`
`Matrix Labs., Ltd., 619 F.3d 1346, 1354 (Fed. Cir. 2010) (emphasis in original).
`
`Petitioner’s selection of an FAA as a “lead compound” for AED
`
`development in 1996 is based on impermissible hindsight. By the 1996 priority
`
`date, not one AED approved by the FDA or in clinical trials had the FAA
`
`backbone found in lacosamide. Ex. 2036 ¶¶197–198, Table 2. The complete lack
`
`of demonstrated clinical efficacy or acceptable side effects for this entire class of
`
`compounds would have deterred a POSA from choosing any FAA compound as a
`
`“natural choice for further development efforts.” Otsuka, 678 F.3d at 1291. See Ex.
`
`2036 ¶¶196–219.
`
`10
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`Instead, a POSA looking for a lead compound from which to develop a new,
`
`safe and effective AED in 1996 would have modified an existing FDA-approved
`
`drug or a compound having proven clinical efficacy. See, e.g., Ex. 2014 at 119,
`
`126. Looking to existing, approved products was a common approach to AED
`
`development at the time of the invention. See id. at 96–97. In fact, approximately
`
`half of all AEDs in 1996 were developed through structural modification of a then-
`
`existing AED. Ex. 2036 ¶¶200–204. For example, phenobarbital was modified to
`
`produce mephobarbital, metharbital, and primidone. See Ex. 2015 at 234.
`
`Phenytoin was modified to produce trimethadione, phensuximide, methsuximide,
`
`ethotoin, and ethosuximide. See id. at 227, 244, 247. Diazepam was modified to
`
`produce clonazepam,
`
`lorazepam and clorazepate. See
`
`id. at 238–241.
`
`Acetazolamide was modified into methazolamide. See Ex. 2036 ¶203.
`
`In contrast, none of the twenty-four AEDs that had been launched in the
`
`United States by the 1996 priority date—not a single AED—contained the FAA
`
`backbone, as depicted below in Fig. 1. See Ex. 2036 ¶¶112, 198.
`
`Fig. 1
`
`
`
`Among all of the researchers working to discover safe and effective
`
`11
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`antiepileptic compounds in 1996, Petitioner cites not one prior art reference from
`
`another source suggesting that FAAs should be investigated. Even among
`
`investigative AED compounds—which were not yet FDA approved and for which
`
`less clinical and pre-clinical data existed—cumulative review articles identifying
`
`promising AED compounds as of 1996 did not mention Dr. Kohn’s FAAs. See,
`
`e.g., Exs. 2014, 2015; Ex. 2036 ¶¶206–207. Indeed, not even Petitioner disputes
`
`that Dr. Kohn was outside the mainstream of AED research.
`
`In addition, a POSA would have specifically avoided selection of a
`
`compound or class of compounds, such as FAAs, for which the mode of action or
`
`target was not understood. Ex. 2036 ¶¶208–209. As of 1996, pharmaceutical
`
`companies considered knowledge of the mode of action to be a critical factor in
`
`determining whether to investigate a particular class of compounds such as AEDs.
`
`Id. ¶209. Without knowledge of the mode of action, the chances of success in
`
`developing such a compound would be very low. Id. ¶208. Lack of knowledge of
`
`the target to which FAAs bind to produce their biological and pharmacodynamic
`
`effects was also a reason to avoid this class of compounds, because development
`
`would essentially consist of the extraordinarily difficult task of designing a key
`
`without knowing the shape of the lock. Id. ¶¶78–79. Discouraging a POSA even
`
`further was the limited data regarding the structure-activity relationships, which
`
`would have made development of this class of compounds even more difficult. Id.
`
`12
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`¶¶211–219. The data available for FAA compounds in 1996 provided scarce
`
`guidance regarding which structures would cause unacceptable neurotoxicity. Id.
`
`¶215. There was also no clear trend as to structure and protective index. Id. ¶216.
`
`In addition, and as Dr. Wang admits, the prior art contained no information on
`
`long-term toxicity of FAAs. See Ex. 2035, 164:8–166:4, 169:9–171:6; see also Ex.
`
`2036 ¶218. A POSA seeking to optimize these important properties would have
`
`been reduced to relying on chance.
`
`Ignoring the true state of the art at the time of the invention of the ’551
`
`patent, the Petition fails to address why a POSA would have had any reason or
`
`motivation to select a compound from the untested class of FAAs. No FAA had
`
`demonstrated efficacy or safety in humans. In sharp contrast, there were thousands
`
`of non-FAA options including, for example, FDA-approved AED structures,
`
`derivatives thereof, and other structurally different AEDs in clinical or advanced
`
`preclinical development, with published pharmacological data, that represented
`
`suitable starting points. See, e.g., Exs. 2014–2023.3
`
`
`3 Exhibits 2016–2017 identify 24 antiepileptic drugs—none of which are FAAs—
`
`marketed in the U.S. prior to the March 1996 priority date. See also Exs. 2018,
`
`2021–2023 (providing chemical structures of these drugs). FDA next approved an
`
`antiepileptic drug, also not an FAA, in August 1996. See Exs. 2019, 2020.
`
`13
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`B.
`
`Petitioner fails to show that a POSA would have selected
`Compound 3l as a lead compound.
`
`Petitioner fails to demonstrate that, even if a POSA would have looked to
`
`FAAs, a POSA would have then selected Compound 3l as a lead compound. Using
`
`the claims as a starting point, Petitioner and Dr. Wang focus only on efficacy data
`
`of one compound in one particular 1991 article (see Pet. at 44, 47–48) while
`
`ignoring other FAA data that would have directed a POSA to differently-structured
`
`and more promising FAA compounds. In fact, the FAA prior art as a whole—prior
`
`art Dr. Wang concedes would have been within the knowledge of a POSA (see Ex.
`
`1002
`
`¶12; Ex.
`
`2035,
`
`62:21–66:9)—taught
`
`that
`
`structurally-different
`
`heteroaromatic compounds were highly potent and had low neurotoxicity.4 A
`
`POSA would have also known that the nitrogen-oxygen bond in Compound 3l was
`
`undesirable, and that by 1996 there was a complete lack of interest in Compound
`
`4 In this response, “heteroaromatic compounds” refers to FAA compounds that
`
`contain a heteroaromatic group at the α-carbon position, which is the R2/R3
`
`position in Fig. 1 above. See Ex. 2036 ¶243. The FAA prior art also taught that
`
`certain groups at other positions (e.g., R1 and the neighboring carbonyl, and R)
`
`showed good activities and neurotoxicity profiles, but these promising groups at
`
`these other positions, like the heteroaromatic group at the α-carbon position, are
`
`not contained in Compound 3l. See id. ¶¶232–241, 264.
`
`14
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`3l. Ex. 2036 ¶¶228–281. For these reasons, a POSA would not have selected
`
`Compound 3l as a lead, and Petitioner and Dr. Wang can only rely on hindsight
`
`and the teachings of the ’551 patent for their selection of Compound 3l. See, e.g.,
`
`Ex. 2035, 39:9–40:4 (Dr. Wang explaining that his objective in reviewing
`
`documents was to “take a look” at the patent and the prior art “and see whether
`
`these prior art would make – would affect the patentability of [the patent]”).
`
`1.
`
`The FAA prior art taught that heteroaromatic FAAs were
`preferred.
`
`The FAA prior art pointed to compounds with heteroaromatic substituents at
`
`the α-carbon as the most promising. Ex. 2036 ¶¶243–250 (citing teachings of
`
`numerous publications—teachings that Petitioner and Dr. Wang ignore). These
`
`heteroaromatic compounds
`
`showed potent anticonvulsant activity,
`
`low
`
`neurotoxicity, and reasonable P.I. (“protective
`
`index”) values. Id. ¶247.
`
`Compounds with heteroaromatic substituents at the α-carbon were repeatedly cited
`
`in the FAA prior art as the most promising compounds right up until 1996. Id. ¶248
`
`(citing Bardel 1994); see, e.g., Ex. 1017 at 3355 (observing that “[t]he
`
`experimental findings provided further documentation of the beneficial properties
`
`gained by incorporation of aromatic groups at the C(α)-site and the importance of
`
`heteroatom location within the aromatic ring system for maximal biological
`
`activity”). For example, a comparison of the stark difference in potencies between
`
`heteroaromatic Compound 4 in Kohn 1993 and its non-aromatic analogs
`
`15
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`Compounds 21a and 21b, also in Kohn 1993, demonstrated the importance of
`
`selecting an FAA with a heteroaromatic group at the α-carbon. Ex. 1017 at 3352
`
`(compare heteroaromatic Compound 4, ED50 = 10.3 mg/kg with non-aromatic
`
`Compound 21a, ED50 = 51.7 mg/kg and non-aromatic Compound 21b, ED50 = 89.8
`
`mg/kg); Ex. 2036 ¶253; see also Ex. 2035, 143:16–145:2 (finding “no reason to
`
`question” the conclusion in Kohn 1993 that “[t]he decreased activity of
`
`[compounds] 21 versus 4 can be attributed to the loss of the aromatic ring at the
`
`alpha carbon site ….”), 147:11–147:18 (admitting that the Wang declaration
`
`contains no data that contradicts the preference for α-aromaticity).5
`
`A POSA would have therefore known that removal of a heteroaromatic
`
`group would result in a corresponding loss of anticonvulsant activity. See also Ex.
`
`2036 ¶¶250, 253. Specifically, a POSA would have known that the compounds
`
`exhibiting the best potency were those compounds with 1) a heteroaromatic group
`
`having a heteroatom two atoms removed from the α-carbon, 2) a heteroaromatic
`
`
`5 Kohn 1993 (Ex. 1017) confirmed findings of Dr. Kohn’s earlier work that a
`
`heteroaromatic group at the α-carbon improved activity. See Ex. 1018 at 919
`
`(“Evidence is presented that placement of a relatively small, electron-rich,
`
`heteroaromatic moiety at the α-site leads to a substantial enhancement in the
`
`anticonvulsant activity of the drug candidate ….”).
`
`16
`
`

`
`IPR2016-00204
`
`029819.0100-US03
`
`group containing at least one nitrogen, and 3) a six-membered heteroaromatic
`
`group having more than one nitrogen. See, e.g., Ex. 1017 at 3351 (compare
`
`Compound 7 with Compound 19), 3354; Ex. 2056 at 4569 (Compounds 11 and
`
`13); Ex. 2036 ¶¶246–248. Because the FAA prior art taught that a heteroaromatic
`
`substituent at the α-carbon was important for potency and low neurotoxicity, a
`
`POSA would not have sel