`
`
`
`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`Case IPR2017-01053
`Patent 8,268,299
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`ARGENTUM PHARMACEUTICALS LLC,
`Petitioner
`
`v.
`
`ALCON RESEARCH, LTD.,
`Patent Owner
`
`
`Case IPR2017-01053
`Patent 8,268,299
`
`
`PATENT OWNER ALCON RESEARCH, LTD.’S RESPONSE
`
`
`
`
`
`
`
`
`
`
`
`

`

`
`
`
`I. 
`
`II. 
`
`
`
`Table of Contents
`
`Case IPR2017-01053
`Patent 8,268,299
`
`Introduction and Summary of Argument ........................................................ 1 
`
`Person of Ordinary Skill in the Art .................................................................. 5 
`
`III.  Argentum Has Failed To Meet Its Burden To Prove Obviousness. ................ 6 
`
`A.  Argentum Has Failed To Prove Obviousness over Xia,
`Schneider, and Chowhan. ...................................................................... 6 
`
`1. 
`
`2. 
`
`3. 
`
`4. 
`
`The POSA Would Not Have Combined Xia, Schneider,
`and Chowhan To Arrive at the Claimed Invention. .................... 6 
`
`The Limitations on Concentrations of Anionic Species
`Would Not Have Been Obvious to the POSA. ......................... 24 
`
`The ’299 Patent’s Propylene Glycol and Sorbitol
`Limitations Would Not Have Been Obvious to the
`POSA. ....................................................................................... 30 
`
`The ’299 Patent’s pH Limitations Would Not Have Been
`Obvious to the POSA. ............................................................... 41 
`
`B. 
`
`C. 
`
`Argentum Has Failed To Prove Obviousness over Xia,
`Schneider, Chowhan, and Gadd. ......................................................... 47 
`
`Argentum Has Failed To Prove Obviousness over Xia,
`Schneider, Chowhan, and the TRAVATAN® Label. .......................... 51 
`
`D.  Argentum Has Failed To Prove Obviousness Over Xia,
`Chowhan, Gadd, and the TRAVATAN® Label. ................................. 52 
`
`IV.  Objective Indicia Demonstrate the Non-Obviousness of the Invention. ....... 54 
`
`A. 
`
`B. 
`
`The Claimed Invention Has Led to the Commercial Success of
`TRAVATAN Z®. ................................................................................. 55 
`
`The Claimed Invention Has Met a Long-Felt Need and Has
`Been Widely Accepted. ....................................................................... 56 
`
`V. 
`
`CONCLUSION .............................................................................................. 58 
`
`
`
`i
`
`

`

`
`
`I.
`
`
`
`Introduction and Summary of Argument
`
`Case IPR2017-01053
`Patent 8,268,299
`
`The invention claimed in U.S. Patent No. 8,268,299 (“the ’299 patent”)
`
`combines particular concentrations of zinc ions, borate, sorbitol, and propylene
`
`glycol to achieve a “self-preserved” composition—i.e., one that has sufficient
`
`antimicrobial activity to pass standard tests for “preservative efficacy” without
`
`needing a conventional preservative.1 Argentum dismisses this invention as an
`
`obvious repackaging of well-known components. But while the ingredients in the
`
`claimed compositions were known, there is nothing obvious about the claimed
`
`combination. To the contrary, the person of ordinary skill in the art (“POSA”)
`
`would not have had reason to select and combine the claimed ingredients, let alone
`
`combine them at the claimed concentrations.
`
`All the grounds at issue in this IPR center on the combination of three
`
`references—Xia, Schneider, and Chowhan—and all suffer from the same threshold
`
`defect. Argentum has failed to show that the POSA would have had reason to
`
`combine those references. Indeed, the obviousness of the ’299 patent over Xia and
`
`a Chowhan continuation patent with the identical specification was fully
`
`considered and rejected by the PTO during examination. Argentum’s assertion
`
`1 As used in the ’299 patent, “self-preserved” compositions “do not contain a
`
`conventional antimicrobial preservative, such as benzalkonium chloride,
`
`polyquaternium-1, chlorite, or hydrogen peroxide.” I.D. 7–8; Pet. 6.
`
`
`
`1
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`that the addition of Schneider to the combination somehow renders the invention
`
`obvious is meritless. The premise of Argentum’s obviousness argument is that the
`
`POSA would have been motivated—as the first in a series of modifications—to
`
`replace the benzalkonium chloride (“BAK”) in Schneider’s formulation with the
`
`zinc disclosed in Xia. But Argentum admits (as it must) that Xia by itself teaches
`
`“multi-dose ophthalmic formulations containing a prostaglandin glaucoma agent
`
`that avoids the use of traditional preservatives, including BAC.”2 Pet. 14. This
`
`admission belies any motivation the POSA would have had to combine Xia with
`
`any other reference. Yet, rather than simply using the self-preserved formulations
`
`disclosed in Xia, Argentum posits that it would have been obvious to the POSA to
`
`combine Xia with Schneider and then drastically alter the resulting Schneider/Xia
`
`formulation’s ingredients based on at least Xia, Schneider, and Chowhan.
`
`Even assuming the POSA combined Xia and Schneider, Argentum cannot
`
`explain why the POSA would have been motivated to modify the Schneider/Xia
`
`formulation. Argentum’s own expert agreed that the POSA would have expected
`
`the Schneider/Xia formulation to pass preservative efficacy standards (“PET”).
`
`The best Argentum can muster is an unsupported assertion that the POSA would
`
`have “optimized” the preservative efficacy of the Schneider/Xia formulation by
`
`reducing its zinc concentration below the levels shown to pass PET in Xia, down to
`
`2 “BAC” and “BAK” are both accepted abbreviations for benzalkonium chloride.
`
`
`
`2
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`the levels in the ’299 patent’s claims. But “optimizing” preservative efficacy
`
`would not have led the POSA to decrease the concentration of zinc; doing so
`
`would have been expected to reduce efficacy or encourage microbial growth,
`
`exactly the opposite of “optimizing” preservative efficacy.
`
`Argentum also fails to show any teaching regarding why the POSA would
`
`substitute the single polyol—mannitol—present in the combined Schneider/Xia
`
`formulation, and replace it with a combination of two polyols—propylene glycol
`
`and sorbitol—each at the particular concentration that Alcon claims. Argentum’s
`
`case is based entirely on hindsight. Although Chowhan lists propylene glycol and
`
`sorbitol among its preferred polyols, it gives no reason to combine polyols, much
`
`less a reason to select or combine these two particular polyols; it states that two
`
`other polyols—mannitol and glycerin—are each more preferred than either
`
`propylene glycol or sorbitol. Lacking any reason for the POSA to select the
`
`claimed combination of polyols at the claimed concentrations, Argentum’s expert
`
`testified that the combination would be obvious because the POSA would “go
`
`through the iterations” of potential polyols and polyol combinations at various
`
`concentrations. But selecting particular concentrations of particular polyols would
`
`have required far more than routine optimization, even if it were clear what the
`
`POSA would be optimizing. There was no suggestion in the art that selecting more
`
`than one polyol, propylene glycol and sorbitol in particular, or any particular
`
`
`
`3
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`concentrations of these two polyols, would optimize anything. The inventors
`
`discovered that borate-polyol complexes can form anions that interfere with the
`
`preservative activity of zinc. Alcon devised a formulation to avoid this
`
`interference. Neither the anion-related interference, nor the inventors’ solution to
`
`it, is taught in the art—and certainly not in Xia, Schneider, or Chowhan.
`
`Beyond the zinc, borate, sorbitol, and propylene glycol limitations—which
`
`would not have been obvious individually, much less combined—Argentum also
`
`fails to show the obviousness of the pH range of 5.5 to 5.9, a range that in the
`
`patented invention solves a precipitation problem the POSA would not have even
`
`recognized. Nothing in the prior art suggests this particular range, and the POSA
`
`would not have wanted to decrease pH below 6.0 (the pH taught by Schneider and
`
`the TRAVATAN® label) because doing so would move further away from the pH
`
`of the eye and risk stinging or other irritation. Again, Argentum’s only recourse is
`
`“optimization,” but optimization of the only parameter cited by Argentum—
`
`antimicrobial activity—actually leads away from the claimed range.
`
`The nonobviousness of the claimed invention is further evidenced by
`
`objective indicia of nonobviousness. Among other indicia discussed below, the
`
`invention at issue is what distinguishes Alcon’s TRAVATAN Z® from an earlier
`
`anti-glaucoma product, TRAVATAN®. TRAVATAN Z® sells better than
`
`TRAVATAN® ever did, even though the two are similarly effective—the
`
`
`
`4
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`difference is attributable to the claimed preservative system’s ability to reduce
`
`irritation and damage to the eye. Indeed, TRAVATAN Z®’s commercial success
`
`has held up even in the face of widespread competition from generic drugs,
`
`including generic travoprost—the active ingredient in TRAVATAN Z®—that are
`
`similarly effective but lack its advantageous preservative system.
`
`For all of these reasons, claims 1–28 of the ’299 patent should be upheld.
`
`II.
`
`Person of Ordinary Skill in the Art
`
`The level of ordinary skill in the art is evaluated as of the time the invention
`
`was made. In re Epstein, 32 F.3d 1559, 1564 n.4 (Fed. Cir. 1994). Argentum does
`
`not dispute that September 21, 2006 was the time of invention. See Pet. 7–8.
`
`The definition of the POSA set forth in Argentum’s Petition, Pet. 7, and in
`
`the Declaration of Erning Xia, Ph.D, Ex. 1002 ¶¶15–17, does not differ materially
`
`from the definition of the POSA set forth in the declaration of Alcon’s expert on
`
`ophthalmic formulations, Dr. Soumyajit Majumdar, Ex. 2023 ¶ 17; Ex. 2025 ¶ 16.
`
`Dr. Xia and Dr. Majumdar agree that the POSA would have had at least the
`
`equivalent of a master’s degree in pharmaceutics, chemistry, or a related field, at
`
`least a few years of experience in the development of ophthalmic formulations, and
`
`either (i) education, training, or experience in the field of microbiology including
`
`antimicrobial activity of pharmaceutical formulations and preservative efficacy
`
`testing, or (ii) the ability to consult with microbiologists with such experience. Ex.
`
`
`
`5
`
`

`

`
`
`
`
`1002 ¶¶ 16–17; Ex. 2023 ¶ 17; Ex. 2025 ¶ 16.
`
`Case IPR2017-01053
`Patent 8,268,299
`
`III. Argentum Has Failed To Meet Its Burden To Prove Obviousness.
`“[A] party seeking to invalidate a patent as obvious must demonstrate . . .
`
`that a skilled artisan would have had reason to combine the teaching of the prior art
`
`references to achieve the claimed invention, and that the skilled artisan would have
`
`had a reasonable expectation of success from doing so.” In re Cyclobenzaprine
`
`Hydrochloride Extended-Release Capsule Patent Litig., 676 F.3d 1063, 1068–69
`
`(Fed. Cir. 2012). It is improper when doing an obviousness analysis to evaluate
`
`the obviousness of individual claim limitations in isolation; rather, the claimed
`
`invention “must be considered as a whole . . . and the claims must be considered in
`
`their entirety.” Rockwell Int’l Corp. v. United States, 147 F.3d 1358, 1364 (Fed.
`
`Cir. 1998). Argentum has failed to follow these principles, and in so doing, has
`
`failed to carry its burden to establish the obviousness of any of the claims of the
`
`’299 patent.
`
`A. Argentum Has Failed To Prove Obviousness over Xia, Schneider,
`and Chowhan.
`1.
`
`The POSA Would Not Have Combined Xia, Schneider, and
`Chowhan To Arrive at the Claimed Invention.
`
`All of the claims of the ’299 patent require a combination of zinc, borate,
`
`
`
`6
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`and polyol, as well as a concentration of zinc ions of 0.4 mM or less.3 Argentum
`
`argues that the POSA would have been motivated to combine a particular
`
`travoprost formulation disclosed in Schneider, Ex. 1007, 9:26–42 (“Formulation
`
`A”4), with the zinc disclosed in Xia and then “optimize” the resulting composition
`
`in part by applying Chowhan’s teachings about borate-polyol complexes, to arrive
`
`at the zinc concentrations recited in the claims of the ’299 patent. See Pet. 13–16.
`
`Argentum is wrong. Because Xia disclosed formulations that could be used as a
`
`vehicle for a prostaglandin analogue and passed PET, the POSA seeking to
`
`develop a BAK-free prostaglandin analogue formulation would not have had
`
`reason to combine Xia with any other reference. But even if the POSA combined
`
`
`3 Claims 1–26 of the ’299 patent specify a maximum concentration of zinc ions of
`
`0.4 mM. See Ex. 1001, 25:31–28:57. Claim 27 contains the limitation “ionized
`
`zinc chloride at a concentration of 0.0025% w/v,” and claim 28 contains the
`
`limitation “zinc chloride ionized in the composition at a concentration of 0.0025%
`
`w/v.” See id., 28:17, 28:39–40. Converted to millimoles, both claim 27 and 28
`
`require a concentration of 0.18 mM zinc ions, which is less than 0.4 mM. Ex.
`
`2023¶¶ 22–23.
`
`4 This is a distinct formulation from the formulation A in column 8, within
`
`Schneider’s Example 1. See Ex. 1007, col.8.
`
`
`
`7
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`Xia with Schneider and was intent on further improving the preservative efficacy
`
`of the resulting combined composition, and even if it attempted to do so by
`
`incorporating Chowhan’s teachings about borate-polyol complexes, it still would
`
`not have arrived at the claimed invention. This is because the POSA seeking to
`
`increase the preservative efficacy of the Schneider/Xia formulation would not have
`
`used a concentration of zinc lower than the lowest concentrations used in Xia’s
`
`working examples due to concerns that it could decrease preservative efficacy.
`
`a.
`
`To show obviousness based on combining prior art references,
`
`Argentum must demonstrate “some reason for the combination other than the
`
`hindsight gleaned from the invention.” Uniroyal, Inc. v. Rudkin-Wiley Corp., 837
`
`F.2d 1044, 1051 (Fed. Cir. 1988); Kinetic Concepts, Inc. v. Smith & Nephew, Inc.,
`
`688 F.3d 1342, 1368 (Fed. Cir. 2012) (disallowing “hindsight reconstruction of
`
`references to reach the claimed invention”).
`
`Argentum argues that the POSA would have combined Schneider and Xia
`
`“to improve Schneider’s ophthalmic formulation containing a glaucoma agent by
`
`removing BAC, a known source of toxicity, discomfort, and irritation to [the] eye.”
`
`Pet. 14; Ex. 1002 ¶ 47. As an initial matter, the problem facing the POSA was
`
`broader than merely developing a BAK-free version of Schneider’s Formulation A,
`
`and the POSA would have been motivated to solve the broader problem of
`
`developing a basic BAK-free multi-use ophthalmic formulation that could be used
`
`
`
`8
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`as a vehicle for a variety of different active ingredients. Ex. 2023 ¶¶ 43–44.
`
`Argentum’s expert, Dr. Erning Xia, agrees (unsurprisingly, given that he did just
`
`that). Ex. 2121, 7:18–23.
`
`Even assuming the POSA’s motivation was limited to formulating a BAK-
`
`free version of Formulation A, Argentum does not explain why the POSA would
`
`select zinc over other preservative options that were less toxic than BAK. The
`
`POSA would have been aware, for example, of polycationic polymers such as
`
`Polyquaternium-1 and Polyquaternium-10; sorbic acid; sodium perborate;
`
`stabilized oxychloro complex; and polyhexamethylene biguanide. Ex. 2023 ¶ 45.
`
`Had the POSA discarded those options in favor of zinc—which had never
`
`before been the sole preservative in a marketed ophthalmic drug—the POSA
`
`would have recognized that Xia, without further combination with any other
`
`reference, already solved the problem facing the POSA. Id. ¶¶ 46; Ex. 2025 ¶ 30.
`
`Xia disclosed zinc-preserved ophthalmic formulations that that could be used as
`
`vehicles for prostaglandin analogues and passed stringent PETs but did not contain
`
`a conventional preservative. Ex. 2023 ¶¶ 29–30. Indeed, Xia teaches two solutions
`
`to the problem faced by the POSA: (1) preservation using only zinc (i.e., self-
`
`preservation); and (2) preservation using zinc and “less than a preservative-
`
`effective amount of a primary preservative agent,” such as a cationic polymer like
`
`Polymer JR. See Ex. 1003, 3–4; Ex. 2121, 30:11–32:21; Ex. 2023 ¶ 29. Either of
`
`
`
`9
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`Xia’s methods, according to Xia’s teachings, would provide adequate preservation
`
`without being toxic, uncomfortable, or irritating to the eye. Ex. 2023 ¶¶ 33–35.
`
`Significantly, Argentum admits that Xia by itself solved the problem.
`
`According to Argentum, the “POSA would have appreciated Xia’s disclosure of
`
`multi-dose ophthalmic formulations containing a prostaglandin glaucoma agent
`
`that avoids the use of traditional preservatives, including BAC.” Pet. 14. Despite
`
`appreciating that Xia disclosed the very type of formulation the POSA sought to
`
`develop, Argentum argues that the POSA would nevertheless have taken the more
`
`complicated and less obvious route of adding Xia’s zinc to Formulation A and
`
`incorporating teachings from Chowhan, rather than simply adding a prostaglandin
`
`analogue like travoprost to one of the compositions disclosed in Xia. Xia teaches
`
`formulations that can be used as a vehicle for prostaglandin analogues like
`
`travoprost. Ex. 1003, 12; Ex. 2023 ¶ 44. Accordingly, because the prior art Xia
`
`reference “independently operate[d] effectively,” the POSA “merely seeking to
`
`create a better [formulation] . . . would have no reason to combine the features” of
`
`this prior art with other art—such as Schneider or Chowhan—addressing the same
`
`problem. Kinetic Concepts, 688 F.3d at 1369; see also Winner Int’l Royalty Corp.
`
`v. Wang, 202 F.3d 1340, 1349 (Fed. Cir. 2000).
`
`Indeed, Argentum’s reason for starting with Formulation A (rather than
`
`Xia’s formulations) betrays the influence of hindsight on its theory. According to
`
`
`
`10
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`Argentum and Dr. Xia, the POSA would have begun with Schneider’s Formulation
`
`A because its formulation was already in the marketplace as TRAVATAN®. Ex.
`
`2121, 18:12–16; Pet. 15. Because Formulation A had therefore already
`
`demonstrated safety and effectiveness and been approved by the FDA, the POSA
`
`“would have retained as much of [Schneider’s] travoprost formulation as feasible.”
`
`Pet. 15; Ex. 2121, 18:12–16. But this purported reasoning rings hollow
`
`considering that Argentum’s theory requires that the POSA eliminate three of the
`
`six remaining ingredients in Formulation A (tromethamine, EDTA, mannitol),
`
`changed the concentration of a fourth (boric acid), and added two new ingredients
`
`in addition to zinc (propylene glycol, sorbitol). Pet. 15–19. These changes make
`
`clear that Argentum’s only real reason for starting with and “retaining” portions of
`
`Formulation A is to gin up an explanation—with the benefit of hindsight—for why
`
`the POSA would have combined Chowan’s borate-polyol complexes with Xia’s
`
`zinc: Xia’s formulations do not contain polyol, but Formulation A does. This
`
`hindsight reconstruction of references cannot prove obviousness. Kinetic
`
`Concepts, 688 F.3d at 1368.
`
`b.
`
`Even assuming the POSA did replace Schneider’s BAK with Xia’s
`
`zinc, the POSA would have expected the resulting “Schneider/Xia formulation” to
`
`be stable, non-toxic, and pass PET. The POSA therefore would not have reason to
`
`modify this formulation based on Chowhan. Ex. 2023 ¶ 47; Ex. 2025 ¶¶ 31–32.
`
`
`
`11
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`Indeed, while Argentum and Dr. Xia state repeatedly that the POSA would
`
`“optimize” the Schneider/Xia formulation, neither Argentum nor Dr. Xia
`
`articulates what attributes of the Schneider/Xia formulation the POSA would have
`
`sought to optimize or why.
`
`The foundational premise of Argentum’s argument, moreover, is that the
`
`POSA “would have retained as much of [Schneider’s] travoprost formulation as
`
`feasible, as this was already an FDA-approved formulation[.]” Pet. 15. Assuming
`
`this to be correct, the POSA—in determining what concentration of zinc to use to
`
`replace Schneider’s BAK—would look to the formulation in Xia, among those
`
`meeting preservative efficacy requirements, that is most similar to Formulation A.
`
`That formulation is Example 8, which, like Formulation A, includes EDTA. Ex.
`
`2023 ¶ 48; Ex. 1007, 9:26–42; Ex. 1003, 22. The POSA would therefore have
`
`been motivated to replace the BAK in Formulation A with the concentration of
`
`zinc in Xia Example 8 (0.050 wt.%). Ex. 2023 ¶ 48. The concentration of zinc in
`
`Example 8 is nearly 10 times the concentration of zinc allowed by the broadest
`
`claims in the ’299 patent. Id.; Ex. 1001, 25:34. The POSA would have expected
`
`that replacing the BAK in Formulation A with 0.050 wt.% zinc would result in a
`
`formulation that would pass PET. Ex. 2025 ¶ 31.
`
`Argentum argues that the POSA—despite desiring to maintain as much of
`
`Formulation A as possible—would have removed the EDTA because it was known
`
`
`
`12
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`as a metal chelator and would interfere with zinc’s antimicrobial activity. Pet. 19.
`
`This argument is entirely hindsight-driven, given that Xia itself—the reference on
`
`which Argentum relies for the use of zinc—teaches that its zinc preservative
`
`compositions “may include chelating or sequestering agents,” cites EDTA as a
`
`“preferred” such agent, and included it in example formulations that passed PET.
`
`Ex. 2023 ¶ 49; Ex. 1003, 11. In addition, Schneider teaches that EDTA is present
`
`because of its antimicrobial activity, which the POSA would infer is at least one of
`
`the reasons it was included in Formulation A. Ex. 2023 ¶ 49; Ex. 1007, 7:14–19.
`
`Xia Example 8 passed PET even though its EDTA concentration was five times the
`
`amount in Formulation A. Because both Xia and Schneider teach towards rather
`
`than away from the use of EDTA, the POSA seeking to combine Xia and
`
`Schneider would not have had any reason to set aside this teaching and reject the
`
`use of EDTA or use a different formulation as a model. Ex. 2023 ¶ 49.
`
`Critically, Dr. Xia agreed that the POSA would have a reasonable
`
`expectation that a hypothetical formulation combining Schneider and Xia, based on
`
`Example 8 and having 0.050 wt.% zinc, would be expected to pass PET without
`
`further modification. Ex. 2121, 59:3–12. This is fatal to Argentum’s case.
`
`Without some reason why the POSA would have expected that modifying such a
`
`formulation or combining it with Chowhan would improve some relevant property
`
`of the formulation, it would not have been obvious to take the steps that would be
`
`
`
`13
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`necessary to arrive at the claimed invention or otherwise modify the Schneider/Xia
`
`formulation based on Chowhan. See Kinetic Concepts, 688 F.3d at 1368. Rather
`
`than identifying a motivation for the POSA to combine those references,
`
`Argentum’s theory is simply an improper hindsight reconstruction of the claimed
`
`invention by locating its components in disparate prior art. See id.; Uniroyal, 837
`
`F.2d at 1051.
`
`c.
`
`Nevertheless, Argentum argues that the POSA would have arrived at
`
`the zinc concentrations claimed in the ’299 patent, which are lower than the
`
`concentrations disclosed in Xia’s working examples, by taking advantage of—and
`
`“optimizing” based on Chowhan—the borate-polyol complexes in the
`
`Schneider/Xia formulation. Pet. 15–17. In so arguing, however, Argentum fails to
`
`point to (i) a reason the POSA would alter the zinc concentrations in Xia’s working
`
`examples; or (ii) an expectation that zinc concentrations below Xia’s working
`
`examples provide sufficient preservative efficacy alone or (iii) in combination with
`
`the borate-polyol complexes taught in Chowhan. In addition, (iv) if the POSA
`
`were motivated to “optimize[] the type and amount of borate-polyol complex as
`
`well as the concentration of zinc to maximize antimicrobial effect,” Ex. 1002 ¶ 51,
`
`the POSA would increase rather than decrease the concentration of zinc.
`
`i.
`
`Argentum does not articulate a reason why the POSA would be
`
`motivated to use a concentration of zinc below the levels disclosed in Xia’s
`
`
`
`14
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`working examples. No reference Argentum cites suggests, for example, that the
`
`levels of zinc in Xia’s examples would cause ocular irritation, nor would the POSA
`
`have thought that they would. Ex. 2023 ¶¶ 33–35, 56. Dr. Xia agreed that there
`
`was nothing undesirable about those zinc concentrations. Ex. 2121, 60:2–19,
`
`69:11–17, 70:23–71:5; Ex. 2023 ¶¶ 33, 35, 56. Indeed, Xia’s central teaching is
`
`that its ophthalmic compositions “ha[ve] the benefit of being adequately preserved
`
`without having a harsh physiological effect such as irritation or discomfort caused
`
`by at least some traditional preservative agents.” Ex. 1003, 4. In addition, the
`
`POSA would have known that zinc salts have been used as ophthalmic astringents
`
`at concentrations of around 0.25% w/v; that is, concentrations many times higher
`
`than the highest concentration of zinc chloride used in any of Xia’s examples. See
`
`Ex. 2032, 7089; Ex. 2023 ¶ 56.
`
`The lowest concentration of zinc disclosed in any of Xia’s examples, or for
`
`which Xia provides any PET data, is 0.0065 wt.%, the amount in Xia’s Example
`
`18. See Ex. 1003, 23; Ex. 2121, 63:14–18. Because Example 18 has no
`
`preservative ingredients other than zinc, yet passes a preservative efficacy test
`
`similar to USP 27, the POSA would understand from Xia that this amount of zinc
`
`suffices to meet preservative efficacy requirements. Ex. 2121, 30:17–31:4; Ex.
`
`2023 ¶¶ 31–32; Ex. 1002 ¶¶ 60–61; Ex. 1003, 21, 23. But this amount of zinc is
`
`larger than the than the maximum amount—0.4 mM—in any claims of the ’299
`
`
`
`15
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`patent, and more than twice the zinc ion concentration in claims 14, 24, 27, and 28;
`
`when dissolved in an aqueous solution, 0.0065 wt.% of zinc chloride results in 0.48
`
`mM of zinc ions. Ex. 2023 ¶¶ 31, 55. In contrast, there are no data in Xia
`
`suggesting that any lower amount of zinc is enough. Id. ¶ 32 Accordingly, the
`
`POSA seeking to create a BAK-free version of Schneider’s Formulation A would
`
`have had reason to avoid a concentration of zinc below 0.48 mM, and would not
`
`have had reason to use a concentration within the ranges recited in any of the
`
`challenged claims. Id. ¶¶ 53–56, 66.
`
`ii.
`
`To escape this conundrum, Argentum focuses on Xia’s generic
`
`disclosure that its invention encompasses use of “a minimum of about 0.001 wt.%”
`
`or “0.005 wt.%”of a zinc compound. Pet. 15–16. Those amounts, assuming (as
`
`Dr. Xia does) that zinc chloride is the compound, equate to about 0.074 mM and
`
`0.37 mM, respectively. Ex. 1002 ¶ 50; Ex. 2023 ¶ 59. A fatal flaw in Argentum’s
`
`argument, however, is that nothing in Xia would suggest to the POSA that these
`
`lower amounts of zinc alone provide sufficient preservative efficacy.
`
`The POSA would recognize that Xia discloses two distinct preservation
`
`methods: using zinc alone, or using zinc in combination with a cationic polymer
`
`such as Polymer JR, which, as Xia discloses, enhances the antimicrobial activity of
`
`zinc compositions. See Ex. 1003, 5. The POSA would have understood that a
`
`combination of zinc and a cationic polymer could provide adequate preservation
`
`
`
`16
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`using a lower concentration of zinc than a formulation that does not have a cationic
`
`polymer. Ex. 2023 ¶¶ 62–64. But the statement upon which Argentum relies is
`
`part of a general description of Xia’s invention, referring to both of the two distinct
`
`preservation methods. Id. ¶¶ 60, 62; Ex. 1003, 5. Thus, the POSA would have no
`
`reason to believe that Xia’s reference to concentrations of zinc as low as 0.001
`
`wt.% and 0.005 wt.% meant that such low concentrations are preservative-effective
`
`by themselves; to the contrary, the POSA would understand the disclosure of those
`
`concentrations to refer to formulations in which zinc’s activity is enhanced by a
`
`cationic polymer. Ex. 2023 ¶¶ 42–44. Xia provides no preservative efficacy data
`
`for such amounts of zinc, and therefore does not substantiate any notion that they
`
`could pass PET on their own, without a primary preservative. Although
`
`formulations containing a primary preservative like Polymer JR are not “self-
`
`preserved,” and are therefore outside the ’299 patent’s claims, Xia teaches no
`
`reason to avoid the use of Polymer JR in combination with low concentrations of
`
`zinc.
`
`Dr. Xia could not identify any reason that the POSA would have had to use
`
`zinc concentrations below those used in Xia’s examples other than a general, not
`
`zinc-specific, “rule of thumb”—not taught in any art cited by Argentum—to use as
`
`little of any excipient as possible. Ex. 2121, 39:8–11; 60:20–61:5. Indeed, Dr. Xia
`
`testified that because Xia contains no data about the preservative efficacy of
`
`
`
`17
`
`

`

`Case IPR2017-01053
`
`
`Patent 8,268,299
`
`
`concentrations of zinc below 0.0065 wt.%, “nobody knows” whether
`
`concentrations below that would pass PET, but that the POSA would try lower
`
`concentrations because they might pass. Id. 63:22–64:6. Dr. Xia’s position
`
`appears to be that as long as the prior art does not foreclose the possibility of doing
`
`something—in this case using lower zinc concentrations—doing that thing would
`
`be obvious. The actual standard, however, is quite different; there must be a
`
`suggestion to do the thing such that there would be an expectation of success.
`
`Cyclobenzaprine, 676 F.3d at 1068–69. Here, the mere possibility that one could
`
`try a lower zinc concentration is a far cry from a suggestion that the POSA should
`
`do so, much less that it should be done with an expectation of success. Argentum
`
`and Dr. Xia wholly fail to provide any reason (other than hindsight knowledge of
`
`Alcon’s claimed invention) why the POSA would use zinc concentrations below
`
`those used in Xia’s already effective formulations.5
`
`
`5 The foregoing discussion explains why the POSA would not have had a reason to
`
`use a zinc concentration below 0.48 mM, and thus would not have had reason to
`
`practice the broadest independent claims, which require a concentration up to 0.4
`
`mM. The POSA would have had even less reason to use a concentration of
`
`0.0025% zinc chloride, as required by claims 14, 24, 27, and 28, which works out
`
`to approximately 0.18 mM. Ex. 2023 ¶ 55.
`
`
`
`18
`
`

`

`
`
`
`Case IPR2017-01053
`
`Patent 8,268,299
`
`Indeed, the POSA would affirmatively be concerned that zinc compositions
`
`with less zinc than the 0.48 mM in Xia’s Example 18 would fail PET. Ex. 2025
`
`¶¶ 34–39. As explained in the declaration of Dr. George Zhanel, the prior art
`
`taught that a 1 mM concentration of zinc was required to eradicate E. coli—one of
`
`the bacteria included in standard PET—and that, as concentrations were lowered,
`
`the antibacterial effect of zinc was reduced—to a partial reduction in survival at
`
`0.5 mM, and to a growth stimulating effect at 0.25 mM. Id. ¶¶ 35–37. The prior
`
`art likewise suggested that even higher concentrations of zinc than employed in
`
`Xia Example 18 were needed to control P. aeruginosa, another species included in
`
`standard PET. Id. ¶ 38. The inability of low concentrations of zinc to control P.
`
`aeruginosa would have given the POSA particular concern about using a zinc
`
`concentration below the levels in Xia’s examples. That is because P. aeruginosa is
`
`a particularly dangerous bacterium in the ophthalmic context, where it can cause
`
`sight-threatening infections. Id.
`
`iii. Argentum argues that the POSA would have recognized that the
`
`Schneider/Xia formulation contained a borate-polyol system within the range
`
`disclosed in Chowhan, and