UNITED STATES PATENT AND TRADEMARK OFFICE
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`INNOPHARMA LICENSING, INC., INNOPHARMA LICENSING LLC,
`INNOPHARMA INC., INNOPHARMA LLC,
`MYLAN PHARMACEUTICALS INC., and MYLAN INC.
`Petitioner,
`
`v.
`
`SENJU PHARMACEUTICAL CO., LTD., BAUSCH & LOMB, INC., and
`BAUSCH & LOMB PHARMA HOLDINGS CORP.
`Patent Owner.
`
`Case IPR20 14-00903
`Patent 8,129,431
`
`DECLARATION OF STEPHEN G. DAVIES, D.Pbil.
`
`PAGE 1 OF 58
`
`IPR2015-01099
`IPR2015-01097
`IPR2015-01100
`IPR2015-01105
`
`SENJU EXHIBIT 2105 I
`INNOPHARMA v SENJU
`IPR20 15-00903
`
`Lupin EX1126
`Page 1
`
`

`
`Table of Contents
`
`I.
`
`II.
`
`BACKGROUND AND QUALIFICATIONS ........................................... .... .. 2
`
`INFORMATION CONSIDERED ................................................................... 4
`
`Ill. LEGAL PRINCIPLES ...................................... ................ ...................... ......... 5
`
`IV. THE '431 PATENl' ......................................................................................... S
`
`V.
`
`SUMMARY OF OPrNIONS ................................................... ...................... ll
`
`VI.
`
`STATEMENT OF OPINIONS EXPRESSED AND BASES AND
`REASONS THEREFOR .............................................................. ................. 1 4
`
`A.
`
`B.
`
`C.
`
`D.
`
`The Level of Skill in the Art ............................................................... 14
`
`Bromfenac is Structurally and Chemically Dissimilar to
`Diclofenac, Ketorolac and Pranlukast ................................................. 14
`
`I.
`
`2.
`
`3.
`
`Comparison of Bromfenac and Diclofenac ......... .. ................... 14
`
`Comparison ofBromfenac and Ketorolac ......... ....................... 22
`
`Comparison of Bromfenac and Pranlukast ............................... 27
`
`Dr. Laskar Has Not Established That a PrecipHate Will Form
`between an NSAID such as Bromfenac aod BAC ............................. .33
`
`Non-ionic Surfactants are Structurally and Chemically Diverse
`and Not Interchangeable ...................................................................... 36
`
`1.
`
`2.
`
`Comparison of Polysorbate 80 and Tyloxapol ........................ .36
`
`Comparison of0ctoxyno19, Octoxynol40 and
`Tyloxapol .................................................................................. 39
`
`E.
`
`Cyclodextrins May Impact the Stability of the Claimed
`Bromfenac Formulations ..................................................................... 46
`
`VII. NON-OBVIOUSNESS OF THE CLAIMS OF THE '431 PATENT ........... 47
`
`VIII. CONCLUSION ................... ................. ................ .......................................... 50
`
`PAGE 2 OF 58
`
`

`
`I, Stephen G. Davies, under penalty of perjury, declare as follows:
`
`1.
`
`1 have been retained by Finnegan, Henderson, Farabow, Garrett &
`
`Dunner, LLP on behalf of Senju Pharmaceutical, Co., Ltd. in connection with this
`
`action as an expert in organic chemistry and medicinal chemistry.
`
`I.
`
`BACKGROUND AND QUALIFICATIONS
`
`2.
`
`I am the Waynflete Professor of Chemistry at the University of
`
`Oxford, Oxford, England. I have been employed teaching chemistry at Oxford
`
`since 1980. From 2006 to 2011, I was Chairman ofthe Department of Chemistry.
`
`In this position, 1 had full responsibility for all teaching, research, financial and
`
`managerial matters in one of the largest chemistry departments in the world. I
`
`have also supervised more than 100 graduate students and 1 00 post-doctoral
`
`fellows in the area of organic, organometallic and medicinal chemistry.
`
`3.
`
`In 1973, I earned a B.A. in Chemistry from the University of Oxford.
`
`In 1975, I earned a D. Phil. in Chemistry from the University of Oxford. In 1980, I
`
`received a D. Sc. in Chemistry from the University of Paris.
`
`4.
`
`Over the course of my career, 1 have been a committee member of
`
`many professional organizations, a list of which can be found in my curriculum
`
`vitae, attached as Exhibit 2049.
`
`5.
`
`I have authored over 550 publications and have given scores of
`
`research lectures. My research interests include synthetic organic and medicinal
`
`PAGE30F58
`
`

`
`chemistry, and in particular, the preparation of enantiomerically pure organic
`
`compounds,
`
`including
`
`the asymmetric and
`
`stereoselective synthesis of
`
`enantiomerically pure organic compounds for potential therapeutic use. A list of
`
`my publications may be found attached as Exhibit 2050.
`
`6.
`
`I have also held several editorial appointments. I am the Founding
`
`Editor and Editor of Organic Series of "Oxford Chemistry Primers" and "Oxford
`
`Chemistry Masters," an Executive Editorial Board Member of "Tetrahedron," the
`
`Founding Editor and Editor-in-chief of "Tetrahedron: Asymmetry," and the Editor
`
`of the "On Chemistry'' Books.
`
`7.
`
`Over the course of my career, I have received several awards,
`
`including the Hickin bottom Fellowship ( 1984 ), Pfizer Award for Chemistry ( 1985),
`
`1984 Corday Morgan Medal, Royal Society of Chemistry (1986), Royal Society of
`
`Chemistry A ward for Organometallic Chemistry (1987), Pfizer A ward for
`
`Chemistry ( 1988), Royal Society of Chemistry Bader A ward ( 1989), Tilden
`
`Lecture Award, Royal Society of Chemistry ( 1996), Royal Society of Chemistry
`
`A ward in Stereochemistry (1997), Prize Lectureship of the Society of Synthetic
`
`Organic Chemistry, Japan (1998), Distinguished Technopreneur Award, Singapore
`
`(2008), and Royal Society of Chemistry Perkin Prize for Organic Chemistry
`
`(2011).
`
`PAGE 4 OF 58
`
`

`
`8.
`
`I am also the founder of numerous companies including ones focused
`
`on the preparation of compounds for potential pharmaceutical use. Along with
`
`several others, I founded Oxford Asymmetry, Ltd. in 1992, which became a
`
`division of Oxford Asymmetry International plc, with a mission to provide
`
`pharmaceutical companies with enantiomerically pure compounds of interest on
`
`any desired scale, from small amounts for biological evaluation and research, to
`
`commercial quantities. Currently, I am the Founder and Non-executive Chairman
`
`of Scilnk Ltd. I am also the Non-executive Director of Isis Innovation Ltd. I am
`
`also the Founder and Non-executive Director of OxStem Ltd.
`
`I am also the
`
`Founder and Non-executive Director of Summit Therapeutics plc, which develops
`
`pharmaceutical compounds and has two such compounds currently undergoing
`
`Phase IT clinical trials.
`
`9.
`
`On the basis of my education and the experience described above, I
`
`believe r am qualified to give the opinion set out herein.
`
`II.
`
`INFORMATION CONSIDERED
`
`10. The opinions expressed in this declaration are based on my review of
`
`U.S. Patent No. 8,129,431 ("the '431 patent"), the "Petition for Inter Partes
`
`Review of U.S. Patent No. 8,129,431" ("Petition") and Exhibits to the Petition,
`
`including the declaration of Dr. Paul A. Laskar (EX1003). I have also reviewed
`
`the transcript of Dr. Laskar's cross examination (EX2114).
`
`I have further
`
`PAGES OF 58
`
`

`
`reviewed the transcript of the deposition of Dr. Jayne Lawrence, who serves as
`
`InnoPharma's expert in the district court litigation involving the '431 patent
`
`(EX2140). I also base this opinion on my professional and academic experience in
`
`the areas of organic chemistry and medicinal chemistry.
`
`I reserve the right to
`
`testify about these materials and experience. As I discuss below, I disagree with
`
`Dr. Laskar's conclusions that the claims of the '431 patent are invalid based on
`
`obviousness.
`
`III. LEGAL PRINCIPLtS
`
`11.
`
`I understand that an obviousness analysis involves a review of the
`
`scope and content of the prior art, the differences between the prior art and the
`
`claimed subject matter, the level of ordinary skill in the art, and objective indicia of
`
`non-obviousness, such as unexpected results, acclaim and commercial success. I ·
`
`understand that for an invention to be regarded as obvious, a person of ordinary
`
`skill in the art must have had a reason to modify the prior art or to combine one or
`
`more prior art references in a manner that would result in the claimed subject
`
`matter with a reasonable expectation of success.
`
`IV. THE '431 PATENT
`
`12.
`
`I understand that InnoPharma has challenged claims 1-22 of U.S.
`
`Patent No. 8, 129,431 ("the '431 patent"), EX1001, in this proceeding. I further
`
`understand that the '43 1 patent has a priority date of January 2 1, 2003.
`
`PAGE60F 58
`
`

`
`13. The '431 patent is directed, generally speaking, to aqueous liquid
`
`preparations consisting essentially of the non-steroidal anti-inflammatory drug
`
`(''NSAID") 2-amino-3-( 4-bromobenzoyl)phenylacetic acid ("bromfenac"), or a
`
`pharmacologically acceptable salt or hydrate thereof, and the non-ionic surfactant
`
`tyloxapol.
`
`(EX1001 at Abstract; claims 1-22.) The '431 patent has two
`
`independent claims (claims 1 and 18) and 20 dependent claims (claims 2-1 7 and
`
`19-22).
`
`14.
`
`Independent claim 1 ofthe '431 patent is directed, generally speaking,
`
`to an aqueous liquid preparation consisting essentially of bromfenac or a
`
`pharmacologically acceptable salt or hydrate thereof and tyloxapol, formulated for
`
`ophthalmic administration, and where when a quaternary ammonium compound is
`
`present, it is benzalkonium chloride ("BAC"). (EX1001 at claim 1.)
`
`15. Dependent claim 2 of the '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 1, where the first component is a
`
`bromfenac sodium salt. (EX1001 at claim 2.)
`
`16. Dependent claim 3 of the '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 1, where the second component is
`
`tyloxapol and the pharmacologically acceptable salt of bromfenac is a sodium salt,
`
`the concentration of tyloxapol is from about 0.01 w/v % to about 0.5 w/v %, the
`
`first component is a bromfenac sodium salt, and the concentration of the
`
`PAGE 7 OF 58
`
`

`
`bromfenac sodium salt is from about 0.01 w/v% to about 0.5 w/v %. (EX1001 at
`
`claim 3.)
`
`17. Dependent claim 4 of the '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 3, where the concentration of tyloxapol
`
`is from about 0.01 w/v % to about 0.3 w/v % and the concentration of the
`
`bromfenac sodium salt is from about 0.05 to about 0.2 w/v %. (EXlOOl at claim
`
`4.)
`
`18. Dependent claim 5 of the '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 4, where the concentration of bromfenac
`
`sodium salt is about 0.1 w/v %. (EXIOOl at claim 5.)
`
`19. Dependent claim 6 of the '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 4, where the concentration of tyloxapol
`
`is about 0.02 w/v %. (EX1001 at claim 6.)
`
`20. Dependent claim 7 of the '43 1 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 1, where the formulation further includes
`
`one or more additives selected from the group consisting of a preservative, buffer,
`
`thickener, stabilizer, chelating agent, and pH controlling agent. (EX1001 at claim
`
`7.)
`
`21. Dependent claim 8 of the '43 1 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 7, where the preservative is BAC, the
`
`PAGES OF 58
`
`

`
`buffer is boric acid and/or sodium borate, the thickener is polyvinylpyrrolidone, the
`
`stabilizer is sodium sulfite, the chelating agent is sodium edetate, and the pH
`
`controlling agent is sodium hydroxide. (EX tOOl at claim 8.)
`
`22. Dependent claim 9 ofthe '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 8, where the pH is from about 7 to about
`
`9. (EXIOOI at claim 9.)
`
`23. Dependent claim 10 ofthe '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 8, where the pH is from about 7.5 to
`
`about 8.5. (EXIOOl at claim 10.)
`
`24. Dependent claim I I of the '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 4, where the concentration ofbromfenac
`
`sodium salt is about 0.2 w/v %. (EXIOOJ at claim ll.)
`
`25. Dependent claim 12 ofthe '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 4, where the concentration of tyloxapol
`
`is about 0.3 w/v %. (EXlOOI at claim 12.)
`
`26. Dependent claim 13 of the '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 12 where the formulation further
`
`includes one or more additives selected from the group consisting of a preservative,
`
`buffer, thickener, stabilizer, chelating agent, and pH controlling agent. (EX I 001 at
`
`claim 13.)
`
`PAGE 9 OF 58
`
`

`
`27. Dependent claim 14 of the '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 13, where said preservative is BAC, said
`
`buffer is boric acid and/or sodium borate, said thickener is polyvinylpyrrolidone,
`
`said stabilizer is sodium sulfite, said chelating agent is sodium edetate, and said pH
`
`controlling agent is sodium hydroxide. (EX1001 at claim 14.)
`
`28. Dependent claim 15 of the '43 1 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 11, where the concentration of tyloxapol
`
`is about 0.02 w/v %. (EXIOOI at claim 15.)
`
`29. Dependent claim 16 ofthe '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 15 where the formulation further
`
`includes one or more additives selected from the group consisting of a preservative,
`
`buffer, thickener, stabilizer, chelating agent, and pH controlling agent. (EXlOOl at
`
`claim 16.)
`
`30. Dependent claim 17 ofthe '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 16, where said preservative is BAC, said
`
`buffer is boric acid and/or sodium borate, said thickener is polyvinylpyrrolidone,
`
`said stabilizer is sodium sulfite, said chelating agent is sodium edetate, and said pH
`
`controlling agent is sodium hydroxide. (EXlOOl at claim 17.)
`
`31.
`
`Independent claim 18 of the '431 patent is directed, generally
`
`speaking, to an aqueous liquid preparation consisting essentially of bromfenac or a
`
`PAGE 100F58
`
`

`
`pharmacologically acceptable salt or hydrate thereof, tyloxapol, boric acid, sodium
`
`tetraborate, EDTA sodium salt, BAC, polyvinylpyrrolidone, and sodium sulfite,
`
`fonnulated for ophthalmic administration, and where BAC is the only quaternary
`
`ammonium compound included. (EX1001 at claim 18.)
`
`32. Dependent claim 19 of the '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 18, where (a) is a bromfenac sodium salt.
`
`(EXlOOI at claim 19.)
`
`33. Dependent claim 20 of the '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 19, where the concentration of
`
`bromfenac sodium salt is from about 0.01 to about 0.5% and the concentration of
`
`tyloxapol is about 0.02 w/v%. (EX1001 at claim 20.)
`
`34. Dependent claim 21 of the '431 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 20, where the concentration of
`
`bromfenac sodium salt is about 0.01 w/v %. (EXIOOI at claim 21.)
`
`35. Dependent claim 22 of the '43 1 patent is directed, generally speaking,
`
`to the aqueous liquid preparation of claim 20 where the concentration of
`
`bromfenac sodium salt is about 0.1 w/v %. (EX 100 l at claim 22.)
`
`PAGE 11 OF 58
`
`

`
`V.
`
`SUMMARY OF OPINIONS
`
`36.
`
`I understand that the Board has granted InnoPharma's petition to
`
`institute this IPR regarding the purported obviousness of claims 1-22 of the '431
`
`patent on the following grounds:
`
`• Groundl: Obviousness ofclaims 1-5,7-14, and 18-19 over U.S.
`
`Patent No. 4,910,225 ("Ogawa") (EX1004) and U.S. Patent No.
`
`6,1 07,34 3 ("Sallmann") (EX l 009)
`
`• Ground 2: Obviousness of claims 6, 15-17, and 20-22 over Ogawa,
`
`Sallmann, and Australian Patent No. AU-B-22042/88
`
`("Fu'')
`
`(EXI 011)
`
`37. As discussed further below, it is my opinion that a person of ordinary
`
`skill in the art would have had no reason to combine the disclosures of Ogawa and
`
`Sallmann, or Ogawa, Sallrnann and Fu, as Dr. Laskar contends, to arrive at the
`
`claimed preparations of the '431 patent. Ogawa teaches the use of sodium sulfite,
`
`a well-known antioxidant (EX2014 at 3:41-55), to chemically stabilize bromfenac
`
`from degradation and prevent the formation of red insoluble matters. (EXI004 at
`
`Experimental Example 6.) 1 Indeed, colored degradation products are typical of
`
`PAGE 12 OF 58
`
`

`
`oxidation reactions. (EX21 04 at 525, 530-31.) A person of ordinary skill in the art
`
`would have readily understood, therefore, that oxidation caused bromfenac's
`
`degradation and that Ogawa solved this problem using sodium sulfite.
`
`(See
`
`EX 1 021 at 5.) There is no teaching in Ogawa of the formation of any salt or any
`
`"complex" between bromfenac and BAC, contrary to Dr. Laskar's argument.
`
`(EX1003 at~ 27.)
`
`of a red insoluble oxidative degradation product.
`
`(EX 1004 at Experimental
`
`Rather, Ogawa discloses only the precipitation
`
`Examples 4-6.)
`
`38.
`
`Sallmann is directed to formulations of diclofenac potassium in
`
`particular and contains no teaching that diclofenac is susceptible to chemical
`
`degradation. (EX 1009 at 1: l - 3:26.) Fu discloses formulations of ketorolac and
`
`BAC and contains no teaching that ketorolac is susceptible to chemical degradation.
`
`(EX lOll at 11 -19.) Fu is directed to preventing the formation of a precipitate in
`
`ketorolac formulations; i.e., physically stabilizing these formulations. (See, e.g.,
`
`EX1011 at Example 5.) Thus, as discussed further below, a person of ordinary
`
`skill in the art would not have been motivated to combine Ogawa and Sallmann, or
`
`Ogawa, Sallmann and Fu, because these references relate to different active
`
`ingredients and provide different solutions to entirely unrelated problems.
`
`PAGE 13 OF 58
`
`

`
`39. Moreover, as discussed further below, even if Dr. Laskar were correct
`
`that a person of ordinary skill in the art would have been motivated to combine
`
`Ogawa, Sal1mann and Fu, which he is not, these references do not disclose any
`
`formulation containing 0.02 w/v% tyloxapol, which is required in claims 6, 15-17
`
`and 20-22 ofthe '431 patent.
`
`Fu discloses formulations containing
`
`0.02% Octoxynol 40 (EX1011 at 18), but, as discussed below, Octoxynol 40 and
`
`tyloxapol are en.tirely different compounds with different structures and different
`
`properties, and thus are not interchangeable. Fu's disclosure of 0.02 w/v%
`
`Octoxynol40 therefore does not in any way teach or suggest 0.02 w/v% tyloxapol.
`
`40. Nor does Sallmann disclose any formulation containing 0.02 w/v%
`
`tyloxapol. With respect to SalJmann, lnnoPharrna and Dr. Laskar argue that "a
`
`person of ordinary skill in the art, when replacing polysorbate 80 with tyloxapol in
`
`Ogawa's Example 6, would have used the concentration of tyloxapol that is
`
`disclosed in Sallmann's Example 2." (Paper 15 at 16 (citing Pet. 19-22, EX1003 at
`
`~~50-51). ) The tyloxapol concentration disclosed in Sallmann's Example 2 is 0.1
`
`w/v% (EX1009 at 8:1 -15), which is five times greater than 0.02 w/v% tyloxapol,
`
`as set forth in claims 6, 15-17 and 20-22 of the '431 patent. Thus, even if Dr.
`
`Laskar were correct that a person of ordinary skill in the art would have combined
`
`Ogawa and Sallmann, which he is not, this combination would not have led a
`
`PAGE 140F58
`
`

`
`person of ordinary skill in the art to make any formulation containing 0.02 w/v%
`
`tyloxapol.
`
`VI. STATEMENT OF OPINIONS EXPRESSED AND BASES AND
`REASONS THEREFOR
`The Level of Skill in the Art
`
`A.
`
`41. As of January 21, 2003, a person of ordinary skill in the art would
`
`have at least a Bachelor's degree in a field such as chemistry, chemical engineering,
`
`pharmaceutical sciences or a related discipline with about three to five years of
`
`work experience in this area, or a comparable level of education and training.
`
`B.
`
`Bromfenac is Structurally and Chemically Dissimilar to
`Diclofenac, Ketorolac and Pranlukast
`
`1.
`
`Comparison of Bromfenac and Diclofenac
`
`42. Ogawa is directed to bromfenac formulations. (EX1004 at 2:45-3: 15.)
`
`Sallmann is directed to diclofenac potassium formulations. (EX1009 at 1 :1-3:26.)
`
`Dr. Laskar argues that a person of ordinary skill in the art would have had a reason
`
`to combine the disclosures of Ogawa and Sallmann.
`
`In particular, Dr. Laskar
`
`argues that "a POSA would have known that substituting bromfenac for diclofenac
`
`would have yielded predictable results because both are NSAIDs with similar
`
`physical and pharmacological properties," citing EX1002. (EX1003 at~ 61-62.) I
`
`disagree. Although EXI 002 correctly identifies bromfenac as a derivative of
`
`amfenac, it incorrectly identifies diclofenac as a di-chloro derivative of amfenac,
`
`PAGE 150F58
`
`

`
`when in fact amfenac and diclofenac have completely different base structures, as
`
`shown in Table 1 below. (EX1002 at 2.)
`
`duYrOH d&HOH
`
`~
`
`..&
`
`~
`
`6f
`
`annnac
`(ININ atrueturw cf brolnfllnlc)
`
`llromfMac
`
`Q
`NH Crroo
`
`9="
`&a~
`
`~
`
`..&
`
`biN atruc1ure Of CllciOfwniC
`
`cfldof'en.lc
`
`Table 1.
`
`Furthermore, EX1002 contains no information regarding the physical
`
`properties of bromfenac and diclofenac. I therefore disagree with Dr. Laskar's
`
`unsupported opinion because, as discussed below, bromfenac and diclofenac are
`
`structurally and chemically dissimilar in numerous· important respects.
`
`43. Table 2 below depicts the chemical structures of bromfenac and
`
`diclofenac.
`
`PAGE 160F58
`
`15
`
`

`
`OH
`
`OH
`
`bromfenac
`
`diclofenac
`
`Br
`
`Table 2.
`
`44. Bromfenac and diclofenac differ significantly in their structure and
`
`three-dimensional shape.
`
`Important differences in
`
`structure between bromfenac and diclofenac include:
`
`•
`
`•
`
`•
`
`•
`
`Bromfenac and diclofenac are derivatives of different base
`structures, as noted above.
`
`Bromfenac is a primary amine (NH2 group), whereas diclofenac
`is a secondary amine (NH group).
`
`Bromfenac has a 4-bromobenzoyl group attached adjacent to
`the NH2 group, whereas diclofenac has a 2,5-dichloropbenyl
`group attached directly to the NH group.
`
`Bromfenac has a carbonyl (C=O) group, whereas d.iclofenac
`does not.
`
`A person of ·ordinary skill in the art would expect these differences to lead to
`
`significantly different functional and chemical properties, as discussed below. I
`
`PAGE 17 OF 58
`
`16
`
`

`
`• 45. For example, the electron density distribution will vary significantly
`
`between bromfenac and diclofenac based on their different chemical structures,
`
`leading to different hydrogen bonding abilities. Specifically, a person of ordinary
`
`skill in the art would expect the different amino groups in bromfenac (NH2) and
`
`diclofenac (NH) to have different basicities and different hydrogen bonding
`
`abilities. (EX2101 at 919.) Moreover, in bromfenac, the aniline fragment bears
`
`only one (benzoyl) electron withdrawing substituent whereas in diclofenac the
`
`aniline fragment bears three (aryl and 2 chlorine atoms) electron withdrawing
`
`substituents, as shown highlighted below.
`
`OH
`
`(I(CI
`
`Y err' NH
`
`, ~
`h-
`
`OH
`
`bromfenec
`
`dlclofenac
`
`Br
`
`46. A person of ordinary skill in the art would also expect the different
`
`distributions of heteroatoms; i.e. , atoms other than carbon, in bromfenac and
`
`PAGE 180F58
`
`17
`
`

`
`diclofenac to lead to different hydrogen bonding abilities. A person of ordinary
`
`skill in the art would expect different hydrogen bonding abilities to result in
`
`different lipophilicities and solubilities in water. (EX2036 at 43-49; EX2040 at 8-
`
`9.)
`
`47.
`
`Bromfenac is described as freely soluble in water in a reference
`
`that Dr. Laskar cites regarding the Bronuck bromfenac sodium ophthalmic solution,
`
`which is formulated at a pH of 8.0-8.6. (EX2039 at 27, right column and 29, left
`
`column.) Moreover, bromfenac's solubility would not be expected to be pH(cid:173)
`
`dependent at moderate pH levels, such as those specified in the '431 patent,
`
`because bromfenac would be expected to be almost entirely in its carboxylate salt
`
`form at pH 6 and higher, as shown in Appendix A. Indeed, InnoPharma's district
`
`court expert Dr. Lawrence candidly admits that the sodium salt of bromfenac is a
`
`water-soluble hydrophilic drug. (EX2 140 at 33:7-9.)
`
`48.
`
`PAGE 190F58
`
`

`
`49.
`
`In fact, bromfenac contains more strong hydrogen bonding sites than
`
`diclofenac, particularly its carbonyl group, as shown in red below:
`
`Br
`
`bromfenac
`
`diclofenac
`
`50. The ability to form strong hydrogen bonds impacts solvation and how
`
`solvated the molecule is in an aqueous solution. Specifically, solvation occurs by
`
`intermolecular interactions, including hydrogen bonding. Because bromfenac has
`
`more strong hydrogen bonding sites than diclofenac, a person of ordinary skill in
`
`the art would expect the bromfenac ion to be better solvated than the diclofenac ion.
`
`A better solvated ion is more likely to stay in solution and less likely to form
`
`PAGE 20 OF 58
`
`19
`
`

`
`insoluble salts or precipitates (EX2072 at 403-04), such as the "complex, that Dr.
`
`Laskar alleges is formed between bromfenac and BAC. (EX1003 at,-r 27.)
`
`51 . Furthermore, a person of ordinary skill in the art would expect
`
`bromfenac's single bromine on its aromatic ring versus diclofenac's two chlorines
`
`on its aromatic ring to impact the polarity of the two molecules. Polarity also
`
`impacts solvation. (EX2099 at 170-7 L.) Because a person of ordinary skill in the
`
`art would expect bromfenac to have more polar regions than diclofenac, he or she
`
`would also expect bromfenac to be less likely to form insoluble salts or precipitates.
`
`(/d.)
`
`52. Bromfenac and diclofenac also have significantly different three(cid:173)
`
`dimensional structures, as shown below. In bromfenac, the brominated aromatic
`
`ring is approximately in the same plane as the rest of the molecule. In diclofenac,
`
`the nitrogen is very sterically hindered, due to the presence of the dual chlorine
`
`atoms, and the chlorinated phenyl group is approximately at a right angle to the
`
`plane of the rest of the molecule. Differences in three-dimensional structure
`
`impact molecular interactions in the solid state. (EX2037 at 659-60.) In solution
`
`such differences also affect the pharmacological properties of molecules, as well as
`
`the solubility of their various salts, as set forth in Lipinski's rules. (EX2036 at 37-
`
`38, 43-49; EX2040 at 8-9.)
`
`PAGE21 OF 58
`
`

`
`.,-
`
`30 Structures ofbromfcnac (left) and diclofenac (right). For bromfenac the 30 co-ordinates
`were generated using MM2 rcfmemeot within the Chem3D package; for diclofenac the 3D co(cid:173)
`ordinates were obtained from X-ray crystal structure data downloaded from the Cambridge
`crystallographic Database.
`
`53. Furthermore, I disagree with Dr. Laskar's statement that "[a] POSA
`
`would have expected that substituting tyloxapol for polysorbate 80 would
`
`successfully, and predictably, solubilize bromfenac and result in a stable
`
`ophthalmic formulation of bromfenac and BAC because tyloxapol and polysorbate
`
`80 had previously been used interchangeably as surfactants in ophthalmic
`
`formulations," citing EX1026 and EX1019. (EX1003 at~ 56.) EX1026 is directed
`
`to emulsions, which are different from solutions such as the aqueous liquid
`
`preparations of the '431 patent.
`
`(EX1026 at 1:5-11.) EX1019 is directed to
`
`tobramycin and diclofenac formulations, not bromfenac formulations. (EX 1019 at
`
`2:38-59.) Neither EX1026 nor EX1019 provides any evidence whatsoever that
`
`bromfenac needs to be or can be solubilized by polysorbate 80 or tyloxapol.
`.
`-
`
`PAGE 22 OF 58
`
`

`
`54.
`
`In fact, as discussed above, because ofbromfenac's hydrogen bonding
`
`ability, a person of ordinary skill in the art would have no reason to assume,
`
`without testing, that bromfenac would form a precipitate with BAC requiring a
`
`solubilizer to prevent. ln addition, the sodium salt of bromfenac is freely water(cid:173)
`
`soluble (EX2039 at 29, left column; EX2140 at 33:1-9), and thus any solubilizing
`
`effect of polysorbate 80 or tyloxapol would not be required to dissolve or
`
`solubilize bromfenac sodium.
`
`Indeed, InnoPharma's district court expert Dr.
`
`Lawrence unequivocally stated, in a peer-reviewed prior art publication, that "it is
`
`no use trying to increase the aqueous solubility of a water-soluble hydrophilic drug
`
`in an aqueous-based surfactant system."
`
`(EX2139 at 423.)
`
`I agree with Dr.
`
`Lawrence that a person of ordinary skill in the art would have had no reason to try
`
`to solubilize an already water-soluble hydrophilic drug such as the sodium salt of
`
`bromfenac by using an aqueous-based surfactant such as tyloxapol.
`
`2.
`
`Comparison of Bromfenac and Ketorolac
`
`55.
`
`I further disagree with Dr. Laskar's opinion that "a POSA would have
`
`further reason to expect to be able to successfully formulate an aqueous liquid
`
`ophthalmic preparation of bromfenac containing tyloxapol and BAC because Fu
`
`discloses stable aqueous preparations ofNSAIDs and BAC containing a surfactant
`
`(Octoxynol 9) that is closely related to (and known to be interchangeable with)
`
`tyloxapol." (EX1003 at~ 64.) Fu discloses formulations containing the specific
`
`PAGE 230F58
`
`

`
`NSAID ketorolac tromethamine, not NSAIDs generally, with Octoxynol 40, not
`
`Octoxynol 9 as Dr. Laskar contends. (EXIOll at 17-19.) As discussed below,
`
`bromfenac and ketorolac are structurally and chemically dissimilar in numerous
`
`respects. Table 3 below depicts the chemical structures of bromfenac and
`
`ketorolac.
`
`Br
`
`Table 3.
`
`OH
`
`OH
`
`bromfenac
`
`ketorolac
`
`56. Bromfenac and ketorolac differ significantly in their structure and
`
`three-dimensional shape.
`
`Important differences m
`
`structure between bromfenac and ketorolac include:
`
`•
`
`•
`
`•
`
`Unlike bromfenac, ketorolac is not a derivative of amfenac.
`
`Bromfenac is a primary amine (NH2 group), whereas ketorolac
`is a tertiary amine (NR3 group).
`
`Bromfenac has a 4-bromobenzoyl group attached adjacent to
`the NH2 group, whereas ketorolac has a simple benzoyl group
`on the aromatic pyrrole group.
`
`23
`
`PAGE240F58
`
`

`
`•
`
`•
`
`Ketorolac has a pyrrole, whereas bromfenac does not.
`
`Bromfenac is an aniline, whereas ketorotac is not.
`
`57. A person of ordinary skill in the art would expect these differences to
`
`lead to significantly different functional and chemical properties, as discussed
`
`below.
`
`II
`58.
`
`For example, the electron density distribution will vary significantly
`
`between bromfenac and ketorolac based on their different chemical structures,
`
`leading to different hydrogen bonding abilities. Specifically, a person of ordinary
`
`skill in the art would expect the amino groups in bromfenac and ketorolac to have
`
`very different hydrogen bonding abilities. (EX2038 at 1 052-55.) In bromfenac,
`
`the ''NH2" group will be able to act as a hydrogen bond donor. In ketorolac, the
`
`nitrogen is neither a hydrogen bond donor nor a hydrogen bond acceptor. A person
`
`of ordinary skil1 in the art would also expect the different heteroatom distributions
`
`in bromfenac and ketorolac to lead to different hydrogen bond acceptor abilities.
`
`A person of ordinary skill in the art would expect different hydrogen bonding
`
`PAGE250F58
`
`

`
`abilities to result in different lipophilicities and solubilities in water. (EX2036 at
`
`43-49; EX2040 at 8-9.)
`
`59.
`
`In fact, bromfenac contains more strong hydrogen bonding sites than
`
`ketorolac, as shown in red below:
`
`Br
`
`bromfenac
`
`ketorolac
`
`60. As discussed above, the ability to form strong hydrogen bonds
`
`impacts solvation and how solvated the molecule is in an aqueous solution.
`
`Specifically, solvation occurs by intermolecular interactions, including hydrogen
`
`bonding. Because bromfenac has more strong hydrogen bonding sites than
`
`ketorolac, a person of ordinary skill in the art would expect the bromfenac ion to
`
`be better solvated than the ketorolac ion. A better solvated ion is more likely to
`
`stay in solution and less likely to form insoluble salts or precipitates. (EX2072 at
`
`403-04.)
`
`61. Moreover, a person of ordinary skill in the art would expect
`
`bromfenac's single bromine on its aromatic ring versus ketorolac's lack of
`
`PAGE260F58
`
`25
`
`

`
`halogenation on its aromatic ring to impact the polarity of the two molecules. In
`
`addition, the benzoyl group in ketorolac cannot interact with its pyrrole nitrogen
`
`atom in the same way as the amine nitrogen atom in bromfenac, which also
`
`impacts polarity. (EX2038 at 1052-55.) Polarity also impacts solvation. (EX2099
`
`at 170-71.) Because a person of ordinary skill in the art would