`_____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`_____________
`MICRO LABS LIMITED AND MICRO LABS USA INC.
`Petitioners,
`
`v.
`
`SANTEN PHARMACEUTICAL CO., LTD. AND ASAHI GLASS CO., LTD.
`Patent Owners.
`_____________
`
`Inter Partes Review No. IPR2017-01434
`U.S. Patent No. 5,886,035
`
`_____________
`
`
`
`SUPPLEMENTAL DECLARATION OF ARON D. ROSE, M.D.
`
`Micro Labs Exhibit 1032
`Micro Labs v. Santen Pharm. and Asahi Glass
`IPR2017-01434
`
`
`
`I, Aron D. Rose, M.D., declare as follows:
`
`I.
`
`INTRODUCTION AND BACKGROUND QUALIFICATIONS
`1.
`I have been retained by the law firm Pillsbury Winthrop Shaw Pittman
`
`LLP (“Pillsbury”) on behalf of Petitioners Micro Labs Limited and Micro Labs
`
`USA Inc. (collectively “Micro”) in connection with the captioned matter.
`
`2.
`
`I previously submitted a declaration in connection with this matter,
`
`marked Micro Labs Exhibit 1028, on May 12, 2017. My background and
`
`qualifications are summarized in my May 12, 2017 Declaration. I also provided a
`
`then-current copy of my curriculum vitae as Appendix B. The most current
`
`version of my curriculum vitae is appended to the present Declaration as Appendix
`
`B.
`
`3.
`
`I understand that Micro filed a petition with the United States Patent
`
`and Trademark Office (“USPTO”) requesting that the USPTO institute a
`
`proceeding to adjudicate the validity of U.S. Patent No. 5,886,035 (“the ’035
`
`patent”). I further understand that Santen Pharmaceutical Co., Ltd. and Asahi
`
`Glass Co., Ltd. (collectively, “Patent Owners”) filed a preliminary response
`
`opposing Micro’s petition. I further understand that the USPTO, having
`
`considered Micro’s petition and Patent Owners’ preliminary response, determined
`
`that Petitioner demonstrates a reasonable likelihood of showing that claims 1–14
`
`
`
`1
`
`Micro Labs Exhibit 1032-2
`
`
`
`would have been obvious over the combined teachings of Klimko, Kishi, and
`
`Ueno.” (Paper No. 11, p. 19.)
`
`II.
`
`SCOPE OF WORK
`4.
`As set forth in my May 12, 2017 Declaration, I am being compensated
`
`at a rate of $500 per hour for my work in connection with my work on this matter.
`
`My compensation is not dependent on the substance of my testimony, my opinion
`
`on any issue, or the outcome of this matter.
`
`5.
`
`I understand that the contents of my Declaration will be relied on in
`
`conjunction with a medicinal chemist to evaluate whether the subject matter
`
`claimed in the ’035 patent would have been obvious to the Person of Ordinary
`
`Skill in the Art as of December 1996 (“POSA”), which is defined in my May 12,
`
`2017 declaration (Ex. 1028).
`
`6.
`
`For the purposes of this Declaration, I have been asked to review the
`
`original and supplemental declarations of Patent Owners’ experts, Exs. 2001, 2002,
`
`2028, and 2029. I have been asked to consider the opinions and analysis expressed
`
`therein, including as it relates to: 1) whether Klimko “teaches away” from selecting
`
`Compound C as a lead compound; 2) whether the POSA would have been
`
`dissuaded or otherwise led away from selecting Compound C as a lead compound
`
`due to concerns about a purported “initial rise” in IOP and/or hyperemia; and 3)
`
`whether the following “secondary considerations” support nonobviousness of the
`
`2
`
`Micro Labs Exhibit 1032-3
`
`
`
`claims of the ’035 patent: commercial success/copying and long-felt but unmet
`
`need.
`
`7.
`
`In preparing this Declaration, I reviewed and considered the
`
`declarations submitted by Dr. Macdonald, marked as Exhibits 2001 and 2028, as
`
`well as the materials referenced therein, and the declarations submitted by Dr.
`
`Fechtner, marked as Exhibits 2002 and 2029, as well as the materials referenced
`
`therein. Their declarations did not change my opinions. As I explain further
`
`below, their declarations contain a number of misleading statements, omissions,
`
`and outright errors.
`
`8.
`
`Further, in addition to the materials I considered for the purpose of
`
`preparing the opinions and analysis contained in my May 12, 2017 Declaration,
`
`and the materials referred to in paragraph 7 above, I have reviewed and considered
`
`those materials listed in Appendix A to this Declaration, which is an updated
`
`version of Appendix to my May 12, 2017 Declaration.
`
`9.
`
`This Declaration summarizes only my current opinions, which are
`
`subject to change depending upon additional information and/or analysis. Either
`
`myself or others working with me or under my direction prepared the exhibits of
`
`the materials that I either reviewed, considered or otherwise relied upon in the
`
`present Declaration. The entirety of my Declaration, including the exhibits and
`
`referenced materials, supplies the basis for my analysis and conclusions. The
`
`3
`
`Micro Labs Exhibit 1032-4
`
`
`
`organizational structure of the Declaration is for convenience. I reserve the right to
`
`supplement my opinions or respond as needed to opinions and assertions made by
`
`others in this matter.
`
`III. LEGAL STANDARDS
`10.
`In addition to the legal standards set forth in my May 12, 2017
`
`Declaration, counsel has informed me of the following legal standards, which I
`
`also apply for the purpose of the present Declaration.
`
`11.
`
`I am informed and understand that evidence of “secondary
`
`considerations” can be relevant to determining whether a claim is obvious.
`
`12.
`
`I am informed and understand that evidence of secondary
`
`considerations of nonobviousness cannot overcome a clear case of obviousness.
`
`13.
`
`I am informed and understand that there must be a nexus between the
`
`merits of the claimed subject matter and any alleged secondary consideration.
`
`14.
`
`I am informed and understand that any alleged secondary
`
`considerations of nonobviousness must be commensurate in scope with the
`
`claimed invention. In other words, for example, evidence of superior properties for
`
`one claimed compound is not necessarily sufficient to support the non-obviousness
`
`of all the other claimed compounds.
`
`4
`
`Micro Labs Exhibit 1032-5
`
`
`
`15.
`
`I am informed and understand that secondary considerations can
`
`include: long-felt and unresolved need, failure of others, unexpected results,
`
`copying, and commercial success.
`
`16.
`
`I am informed and understand that the presence of a long felt and
`
`unresolved need in the art, that is fulfilled by the claimed subject matter, might
`
`weigh in favor of non-obviousness.
`
`17.
`
`I am informed and understand that, in some circumstances, evidence
`
`of others’ attempt, and failure, to develop subject matter similar to the claims can
`
`weigh in favor of non-obviousness.
`
`18.
`
`I am informed and understand that, in some circumstances, evidence
`
`of a superior property, which the POSA would have found surprising or
`
`unexpected at the time of the patent, might weigh in favor of non-obviousness. I
`
`further know and understand that evidence of unexpected results must compare the
`
`claimed invention to the closest prior art. I further know and understand that the
`
`closest prior art is the prior art that is closest to the claimed invention.
`
`19.
`
`I am informed and understand that, when a commercial embodiment
`
`of the claimed subject matter is commercially successful, and there is a nexus
`
`between the commercial success and the claimed subject matter, this factor might
`
`weigh in favor of nonobviousness.
`
`5
`
`Micro Labs Exhibit 1032-6
`
`
`
`20.
`
`I am informed and understand that the fact that others copied the
`
`claimed invention might weigh in favor of non-obviousness. I further understand
`
`that in a case such as this one, where a generic drug company files an application
`
`to seek to obtain FDA approval to market a generic version (“copy”) of the
`
`patented drug, this fact alone is insufficient evidence of non-obviousness. I agree
`
`because I am aware that there are other incentives that must be considered as to
`
`why a generic drug company has sought such FDA approval. For example, I
`
`understand that in some cases, such as here, a generic drug company that is the first
`
`to file an Abbreviated New Drug Application (ANDA), and successfully
`
`challenges the validity of the patent for the branded drug, could be entitled to
`
`receive 180 days of generic marketing exclusivity. In addition, I understand that a
`
`generic company need not perform any marketing to sell its generic drug product if
`
`approved by the FDA because a prescription for the branded product will be
`
`automatically substituted for the generic product at the pharmacy. As a physician,
`
`I am familiar with how generic drugs may be switched for branded drugs at the
`
`pharmacy.
`
`IV. STATE OF THE ART AT THE TIME OF THE ’035 PATENT
`21.
`I described the state of the art in my May 12, 2017 Declaration, Micro
`
`Labs Exhibit 1028 (“Ex. 1028”). In this section I address statements by Dr.
`
`Fechtner in his November 5, 2017 Declaration (“Ex. 2002”).
`
`6
`
`Micro Labs Exhibit 1032-7
`
`
`
`22.
`
`I disagree with Dr. Fechtner’s broad, generalized characterization of
`
`prostaglandins and its early analogs as “clinically unacceptable as a treatment for
`
`glaucoma or ocular hypertension.” (Declaration of Robert D. Fechtner, M.D. (“Ex.
`
`2002”) at 7, ¶ 18.)
`
`23. First, Dr. Fechtner’s discussion of literature from the 1980s glosses
`
`over the considerable progress researchers made between the 1980s and 1996. But
`
`even in the 1980s researchers recognized the clinical potential of prostaglandins
`
`(particularly PGF2α) to lower IOP, in spite of side effects such as hyperemia, ocular
`
`irritation, and/or an initial rise in IOP. (Giuffre, Giuseppe, Graefe’s Arch. Clin.
`
`Exp. Opthalmol. 222:139–141 (1985) at 139 (Ex. 2011 at 1) (“[PGF2α] could be
`
`useful in the treatment of ocular hypertension, although its usefulness would be
`
`limited by the side effects.”).) As early as 1987, one author noted that “early
`
`results on the ocular effects of PGs, which were obtained mostly from rabbits,
`
`should not be generalized; that these reports tended to overstate the adverse effects
`
`of PGs on the eye; and that the primary ocular effect of PGs in most species,
`
`especially primates, is a reduction, rather than an increase in IOP.” (Bito, L.Z. and
`
`Baroody, R.A., “The Ocular Pharmacokinetics of Eicosanoids and their
`
`Derivatives. 1. Comparison of Ocular Eidosanoid Penetration and Distribution
`
`Following the Topical Application of PGF2α, PGF2α-1-methyl ester, and PGF2α-1-
`
`isopropyl ester,” Exp. Eye Res., 44, 217–226 (1987) at 217 (Ex. 2012 at 1). Bito
`
`7
`
`Micro Labs Exhibit 1032-8
`
`
`
`goes on to say that “[i]n fact, PGs appear to be ideally suited for the treatment of
`
`glaucoma by topical application.” Id. at 2.
`
`24. The known side effects of prostaglandins did not dissuade researchers
`
`from pursuing PGF2α analogs even in the 1980s. To the contrary, within a few
`
`years after publication of Giuffre’s work in 1985, there were several papers
`
`published showing a superior therapeutic profile for the isopropyl ester modified
`
`form of PGF2α (“PGF2α-IE”). (Kerstetter, J.R. et al., “Prostaglandin F2α-1-
`
`Isopropylester Lowers Intraocular Pressure Without Decreasing Aqueous Humor
`
`Flow,” American Journal of Ophthalmology, Vol. 105, No. 1, 30–34 (1988) (Ex.
`
`1033); Villumsen and Alm, “Prostaglandin F2α-isopropylester eye drops: effects in
`
`normal human eyes,” Br. J. Ophthalmol. 73:419–26 (Ex. 2013); Villumsen, Jorgen
`
`et al., “Prostaglandin F2α-isopropylester eye drops: effect on intraocular pressure in
`
`open-angle glaucoma,” Br. J. Opthalmol. 73:975–979 (1989) (Ex. 1034). Then,
`
`researchers further improved on the therapeutic profile of PGF2α-IE by modifying
`
`the terminal omega chain with a bulky aromatic substituent, such as phenyl or
`
`phenoxy. (See, e.g., U.S. Patent No. 5,296,504 to Stjernschantz et al. (Ex. 1035);
`
`EP 0 639 563 to Klimko et al. (“Klimko,” Ex. 1003).) One drug that arose out of
`
`that research is the compound latanoprost, approved in June 1996 as the drug
`
`Xalatan. (Ex. 1016; Orange Book Entry for Latanoprost.) Xalatan quickly became
`
`the new gold standard medicinal option for treating elevated IOP. In my own
`
`8
`
`Micro Labs Exhibit 1032-9
`
`
`
`practice, I prescribe Xalatan more frequently than any other medication for the
`
`purpose of treating elevated IOP. In fact, even today, Xalatan accounts for the
`
`majority of prescriptions for treatment of glaucoma or ocular hypertension. (See,
`
`e.g., Alm, Albert, “Latanoprost in the treatment of glaucoma,” Clin. Opthalmol.,
`
`8:1967–85 at 1967 (2014) (Ex. 1036 at 1).)
`
`25. Second, I disagree with Dr. Fechtner’s characterization of PGF2α and
`
`its early analogs as “clinically unacceptable.” Dr. Fechtner’s broad,
`
`uncontextualized declaration is potentially misleading because it oversimplifies the
`
`real process of clinically assessing a potential medication. Clinical assessment of a
`
`potential treatment requires weighing its benefits and potential side effects against
`
`the unique circumstances and needs of a particular patient, considered in the
`
`context of available therapeutic alternatives. The clinician will ultimately
`
`recommend the treatment that he or she concludes is the best option for a particular
`
`patient. Among the remaining treatment options, the clinician could recommend
`
`any one of them under different circumstances, even for the same patient. Dr.
`
`Fechtner’s statement lacks important context and ignores the nuance involved in
`
`clinical analysis.
`
`26. Third, Dr. Fechtner’s opinion is not consistent with my understanding
`
`of the literature. Published data from the 1980s suggests that PGF2α would have
`
`significant clinical efficacy for lowering IOP, and the reported side effects would
`
`9
`
`Micro Labs Exhibit 1032-10
`
`
`
`not be considered intolerable. Accordingly, I would not describe PGF2α, or its
`
`early analogs, as “clinically unacceptable,” and nothing I have reviewed for this
`
`case suggests otherwise. Moreover, the entire history of prostaglandin research
`
`belies Dr. Fechtner’s opinion that “there were concerns that administration of
`
`prostaglandin analogs could result in paradoxical IOP elevations. . . . These same
`
`concerns existed as of December 26, 1996, and a POSA would have recognized
`
`that a candidate prostaglandin analog with paradoxical IOP elevation and/or severe
`
`side effects would not be useful in the clinic.” (Ex. 2002 at 7, ¶ 19.) Specifically
`
`to this point, despite early recognition of the tendency for prostaglandins to exhibit
`
`an “initial increase” in IOP, researchers readily and successfully modified PGF2α to
`
`improve its therapeutic profile. (Kerstetter, J.R. et al., “Prostaglandin F2α-1-
`
`Isopropylester Lowers Intraocular Pressure Without Decreasing Aqueous Humor
`
`Flow,” American Journal of Ophthalmology, Vol. 105, No. 1, 30–34 (1988) (Ex.
`
`1033); Villumsen and Alm, “Prostaglandin F2α-isopropylester eye drops: effects in
`
`normal human eyes,” Br. J. Ophthalmol. 73:419–26 (Ex. 2013); Villumsen, Jorgen
`
`et al., “Prostaglandin F2α-isopropylester eye drops: effect on intraocular pressure in
`
`open-angle glaucoma,” Br. J. Opthalmol. 73:975–979 (1989) (Ex. 1034).)
`
`27. For example, far from being a “serious flaw for a potential IOP-
`
`lowering drug,” as Dr. MacDonald asserts (Ex. 2028 at 2, ¶ 7), modifications of
`
`PGF2α led to latanoprost, a successful prostaglandin analog approved by the FDA
`
`10
`
`Micro Labs Exhibit 1032-11
`
`
`
`in June 1996 to treat patients with elevated IOP. (Ex. 1016.) There was not
`
`skepticism about the clinical usefulness of prostaglandins. To the contrary,
`
`ophthalmologists immediately recognized the clinical potential of latanoprost,
`
`including its advantages over the first-line therapy available at that time (timolol).
`
`(See, e.g., Alan Robin, MD, “Two new options will make glaucoma therapy safer,”
`
`Primary Care Optometry News, July 1996 (Ex. 1037).)
`
`28. As such, I also disagree with Dr. Fechtner’s opinion that, at the time
`
`of the ’035 Patent, ophthalmologists did not believe that prostaglandin analogs
`
`could be used clinically for the treatment of ocular hypertension or glaucoma. (Ex.
`
`2002 at 7, ¶ 19.)
`
`V. KLIMKO DOES NOT TEACH AWAY FROM COMPOUND C
`29. Patent Owners’ experts Dr. MacDonald and Dr. Fechtner opine that
`
`Klimko teaches away from Compound C as a suitable lead compound.
`
`(Declaration of Timothy L. MacDonald (“Ex. 2001”) at 5, ¶ 14; Ex. 2002 at 5, ¶
`
`15.) Broadly speaking, their opinions attempt to rest on two categories of
`
`evidence: 1) certain portions of the data disclosed in Klimko; and 2) how Klimko
`
`characterizes their own data and the data in EP 0 364 417 to Stjernschantz (“Ex.
`
`2017” or “Stjernschantz”). Patent Owners’ experts do not present a full discussion
`
`of Klimko’s data and statements in context. To the contrary, as I explain below,
`
`Dr. MacDonald and Dr. Fechtner present cherry-picked data taken out of context,
`
`11
`
`Micro Labs Exhibit 1032-12
`
`
`
`and uncritically credit Klimko’s opinions without considering their context. Based
`
`on a fulsome review of Klimko by the POSA, in the context of other relevant prior
`
`art, Klimko would not teach away from selecting Compound C as a lead
`
`compound, but in fact would lead the POSA to select Compound C.
`
`A. Klimko’s characterization of the data in Stjernschantz would not
`have deterred the POSA from selecting Compound C.
`30. Dr. MacDonald and Dr. Fechtner, relying on Klimko’s reference to
`
`Stjernschantz, opine that the data in Stjernschantz reveals an “unacceptable
`
`therapeutic profile” for Compound C. (Ex. 2001 at 37–38, ¶ 74; Ex. 2002 at 10, ¶
`
`25.) That characterization is not supported by the data disclosed in Stjernschantz,
`
`nor is it consistent with how Stjernschantz describes his own data.
`
`1.
`
`Patent Owners’ experts mischaracterize the data in
`Stjernschantz.
`31. Dr. Fechtner refers to a supposed “high degree of hyperemia
`
`(2.3±0.3)” for Compound C, relying on a comparison to PGF2α-IE (2.8±0.8), as
`
`shown in Table 4 of Sjernschantz. Id. First, as I explain below in ¶ 44, moderate
`
`hyperemia is a less concerning side effect to the POSA, because it is mainly
`
`cosmetic and wears off within a few hours, so patient concerns can be addressed by
`
`dosing only once before bed. Second, a score of 2.3 (out of 4) is quite modest, and
`
`far from “unacceptable.” The Stjernschantz reference confirms this, describing a
`
`score of 2.0±0.3 for compound 2 (compound D in Klimko) as “only modest
`
`12
`
`Micro Labs Exhibit 1032-13
`
`
`
`conjunctival/episcleral hyperemia.” (Ex. 2017 at 10:57–58.) Third, Dr. Fechtner
`
`did not point out that the administered dose of Compound C (compound 4 in
`
`Stjernschantz) is five times the administered dose for PGF2α-IE. (Ex. 2017 at
`
`15:12–13, 40–41.) This is important because of a well-known principle referred to
`
`as a “therapeutic ceiling.” The therapeutic ceiling for a given drug is the dose at
`
`which, if a higher dose is given, there is no clinically significant increase in
`
`therapeutic effect. But the severity of undesirable side effects can increase as the
`
`dose is increased above the therapeutic ceiling. As such, Dr. Fechtner’s
`
`comparison has little, if any, relevance to determining whether Compound C
`
`exhibits hyperemia to an extent that is comparable to PGF2α-IE, or to any clinically
`
`relevant extent at all.
`
`32. Likewise, the IOP response over time studies reported in Table 5 of
`
`Stjernschantz used a dose of 5 µg for compound C (compound 4 in Stjernschantz).
`
`(Ex. 2017 at 17.) This dose is much higher, over 1600% higher, than the 0.3 µg
`
`effective dose used in Klimko. (Ex. 1003 at 18.) Despite the lower dose used in
`
`Klimko, the IOP-lowering efficacy at the six-hour mark was comparable to
`
`Stjernschantz (about 27% IOP reduction over baseline in Stjernschantz and 28.9%
`
`for Klimko). This strongly suggests that the 5 µg dose used in Stjernschantz is far
`
`above the therapeutic ceiling. The POSA would therefore reasonably expect that
`
`lowering the dose would reduce undesirable side effects, including any initial
`
`13
`
`Micro Labs Exhibit 1032-14
`
`
`
`increase in IOP, without loss of clinical efficacy. This is consistent with literature
`
`published as early as 1977, which shows that an initial rise in IOP was dose-
`
`dependent, and thus could be mitigated by lowering the dose. (See, e.g., Ex. 2003
`
`at Abstract (“Topically applied prostaglandins (PG’s) in the dose range of 25 to
`
`200 µg/eye caused a prolonged (15 to 20 hr.) ocular hypotony (as much as 7 mm.
`
`Hg below control values) in the conscious rabbit following the well-known initial
`
`hypertensive phase. This biphasic intraocular pressure (IOP) response was dose-
`
`dependent and was also observed after intravitreal PG administration. A reduction
`
`of IOP (as much as 7 mm. Hg) lasting for 12 hr. or more was observed following
`
`the topical application of a very low dose (5 µg) of PGF2α which was insufficient to
`
`cause an initial increase in IOP. . . . The present experiments suggest that
`
`exogenous administration of low doses of certain PG’s or their analogues may aid
`
`in the treatment of ocular hypertension”).)
`
`33. Likewise, in contrast to Patent Owners’ experts, Stjernschantz
`
`recognized—consistent with my own opinion and analysis—the importance of the
`
`amount of a particular dose used, particularly when assessing efficacy and side
`
`effects. For example, Stjernschantz discusses the differing doses used in the
`
`studies disclosed therein, noting that “[i]t is clear that with increasing dose side
`
`effects may increase.” (Ex. 2017 at 11.) The failure of Patent Owners’ experts to
`
`recognize this undermines their credibility and their opinions and analysis.
`
`14
`
`Micro Labs Exhibit 1032-15
`
`
`
`2.
`Compound C is characterized favorably by Stjernschantz.
`34. Nowhere in the Stjernschantz reference do the authors characterize
`
`Compound C (compound 4 of Stjernschantz) as having an “unacceptable
`
`therapeutic profile.” To the contrary, the authors of Stjernschantz explicitly
`
`compliment Compound C. For example, on page 11 the author, referring to the
`
`data in Table V, points out that Compound C “effectively lowers the intraocular
`
`pressure as studied in cats.” (Ex. 2017 at 11.) Also, referring to Table III of
`
`Stjernschantz, the author points out that the 16-phenoxy form (i.e. Compound C of
`
`Klimko) “attenuates or abolishes the irritating effect.” Neither Dr. Fechtner nor
`
`Dr. MacDonald even mention Table III of Stjernschantz.
`
`35.
`
`I have reviewed the data in Table III of Stjernschantz and I agree with
`
`Stjernschantz’s statement about the 16-phenoxy form (Compound C of Klimko).
`
`Table III of Stjernschantz discloses a relatively low degree of ocular irritation
`
`(0.3±0.2 for Compound C versus 3.0 for PGF2α-IE), despite a much higher dose (5
`
`µg for Compound C versus only 1 µg for PGF2α-IE). Further, ocular irritation is a
`
`more important side effect to minimize than hyperemia, because the latter is
`
`mainly cosmetic and wears off within a few hours, so patient concerns regarding
`
`hyperemia can be addressed by dosing only once before bed, as I explain above in
`
`¶ 44. Accordingly, the data in Table III of Stjernschantz suggests to the POSA that
`
`Compound C would have a favorable side effect profile.
`
`15
`
`Micro Labs Exhibit 1032-16
`
`
`
`B.
`
`The data in Klimko does not show an undesirable “initial
`increase” in IOP for Compound C.
`36. Dr. Fechtner opines that the data in Figure 2 and Table 4 of Klimko
`
`“confirms the undesirable initial increase in IOP after administration that had been
`
`previously reported in Stjernschantz.” (Ex. 2002 at 14, ¶ 31.) Dr. MacDonald also
`
`opines that Compound C in Klimko exhibits an “initial increase in IOP.” (Ex.
`
`2028 at 2, ¶ 7.) As I explain below, this is incorrect because Klimko did not even
`
`measure the “initial” IOP, i.e. the IOP within a few hours of the first dose, and
`
`further because none of Klimko’s data shows an increase in IOP above baseline.
`
`37. Table 4 of Klimko discloses IOP-lowering activity over time in
`
`response to a 0.3 µg dose of six compounds labeled A, B, C, D, and E. (Ex. 1003
`
`at 18.) Each compound was dosed at the following times: hour zero (initial dose),
`
`hour 8, hour 24, hour 32, and hour 48. In other words, the time interval between
`
`the first four doses alternates between eight hours and sixteen hours. The first
`
`measure of intraocular pressure is at hour 48 (around the same time as the fifth
`
`dose and sixteen hours after the most recent dose). The next three measurements
`
`are at hours 50, 52, and 54 after the initial dose (i.e., 2 hours, 4 hours, and 6 hours
`
`after the fourth dose, respectively). The data in Table 4 is graphically depicted in
`
`Figure 2:
`
`16
`
`Micro Labs Exhibit 1032-17
`
`
`
`Ex. 1003-30
`
`
`
`38. For Compounds A1, B2, and D3, IOP increases toward its pre-dose
`
`level between four and six hours after the most recent dose. This suggests to the
`
`POSA that the IOP lowering effect of each of Compounds A, B, and D begins to
`
`wear off within six hours. In contrast, during the same timeframe the IOP of
`
`Compound C4 is decreasing to its pre-dose level. This suggests to the POSA that
`
`Compound C continues to lower IOP more than six hours after dosing, in contrast
`
`
`1 Cloprostenol
`2 Fluprostenol
`3 17-phenyl-PGF2α
`4 16-phenoxy-PGF2α
`
`17
`
`Micro Labs Exhibit 1032-18
`
`
`
`to compounds A, B, and D. In other words, the trajectory of the data is one feature
`
`that distinguishes Compound C from A, B, and D.
`
`39. Patent Owners’ reference to this data as showing an “initial increase
`
`in IOP” for Compound C, because of the IOP data at hours 0, 2, and 4 after the
`
`fifth dose, is mistaken. This is because, in the study of prostaglandins for lowering
`
`IOP, an “initial increase in IOP” refers to a (usually small) increase in IOP above
`
`baseline, which occurs a few hours after the first dose of a prostaglandin. (See,
`
`e.g., Ex. 2003 (“Camras 1977”) at 4–5; Ex. 1009-0002 (“Camras 1981”) at 2.)
`
`This could be a concern because elevation of IOP above baseline could exacerbate
`
`the patient’s condition. But Klimko does not disclose the IOP measured after the
`
`first (or “initial) dose of compound C. (Ex. 1003 at 18, 30.) Instead, the earliest
`
`IOP measurements disclosed in Klimko for each of Compounds A, B, C, and D,
`
`occur 16 hours after the fourth dose (around the same time as the fifth dose), and
`
`are below baseline to a clinically significant extent. Moreover, following
`
`administration of the fifth dose, none of the IOP measurements (at time = 2, 4 and
`
`6 hours) are above baseline. Indeed, all of the IOP data for Compounds A, B, C,
`
`and D in Figure 2 and Table IV of Klimko are substantially below baseline, much
`
`more below baseline than the commercially-available drug latanoprost.
`
`Accordingly, one cannot say that the data in Klimko says anything about whether
`
`there is an initial increase, much less “confirms” such an effect.
`
`18
`
`Micro Labs Exhibit 1032-19
`
`
`
`40. This is a critical distinction because it completely alters how one
`
`interprets the data in Figure 2 and Table 4 of Klimko. For Compound C, the
`
`reduced IOP-lowering at each time after the fifth dose (0, 2, and 4 hours after)
`
`suggests that the effect of the fourth dose is wearing off, which is precisely what
`
`the POSA would expect if Compound C exhibits longer-lasting IOP lowering
`
`efficacy. The increased IOP-lowering at six hours after the fifth dose suggests that
`
`the clinical effect of the fifth dose is being observed as the effects of the fourth
`
`dose are wearing off. Such a delay is consistent with longer-lasting
`
`pharmacodynamic action, and thus a longer duration of action, which is desirable
`
`for treating elevated IOP.
`
`41. Further, Dr. Fechtner’s and Dr. MacDonald’s assertion that an “initial
`
`increase” in IOP would deter the POSA from pursuing an IOP-lowering drug is
`
`inconsistent with the entire history of prostaglandin research. Despite early
`
`recognition of the tendency for prostaglandins such as PGF2α to exhibit an “initial
`
`increase” in IOP, researchers readily and successfully modified PGF2α to improve
`
`its therapeutic profile. (See, e.g., Ex. 1033; Ex. 2013; Ex. 1034.) Far from being a
`
`“serious flaw for a potential IOP-lowering drug,” as Dr. MacDonald asserts (Ex.
`
`2028 at 2, ¶ 7), further modifications of PGF2α led to latanoprost, a very successful
`
`prostaglandin analog approved by the FDA in June 1996 to treat patients with
`
`elevated IOP. (Ex. 1016.) Accordingly, even if Klimko did suggest an “initial
`
`19
`
`Micro Labs Exhibit 1032-20
`
`
`
`increase in IOP,” which it clearly does not, that would not deter the POSA from
`
`selecting Compound C as a lead compound.
`
`C. Klimko’s hyperemia data would not have deterred the POSA
`from selecting Compound C as a lead compound for further
`development.
`42. Relying on the data in Table 3 and Figure 1 of Klimko, Dr. Fechtner
`
`erroneously opines that the degree of hyperemia exhibited by Compound C would
`
`have eliminated it from further consideration as a lead compound. (Ex. 2002 at
`
`12–13.) I disagree, for several reasons.
`
`43. First, the POSA would have concerns about the reliability of Klimko’s
`
`hyperemia data. For example, Compound B (fluprostenol) exhibits less severe
`
`hyperemia at the higher dose of 0.1 µg as compared to the lower dose of 0.03 µg,
`
`which would not make sense.5 Further, at the higher 1.0 µg dose, Compound B
`
`exhibits less severe hyperemia as compared to the lower dose of 0.3 µg, which
`
`would also make no sense.6 These anomalous results suggest to the POSA that the
`
`experimental method may not have been entirely reliable.
`
`
`5 The number of “2” scores decreased by 13, whereas the number of “1” scores
`increased by 18 and the number of “0” scores decreased by 5. There were no “3”
`scores.
`6 There was a reduction from 29 to 13 for “2” scores, i.e. a change of 16. The
`number of “3” scores increased by only two, whereas the number of “0” and “1”
`scores increased by 14.
`
`20
`
`Micro Labs Exhibit 1032-21
`
`
`
`44. Second, the POSA’s concerns about hyperemia would be reduced
`
`based on the knowledge that the drug could be dosed once at night, before bed.
`
`This is because modest hyperemia, which is a clinical term for redness of the eye,
`
`is considered a cosmetic side effect, and typically only lasts a few hours after
`
`dosing. (See Alan Robin, M.D. “Two new options will make glaucoma therapy
`
`safer,” Primary Care Optometry News (July 1996) at 3 (Ex. 1037 at 3) (“because
`
`it’s given once a day – at nighttime . . . the patient is sleeping during most of the
`
`time the eye is red.”).)
`
`45. Third, I understand from the Kishi reference and from Dr. deLong’s
`
`original declaration that the POSA would reasonably expect that hyperemia would
`
`be reduced by lowering the dose and/or by chemical modification of what I
`
`understand is the C-15 hydroxyl. (Ex. 1005, 1:66–2:15; 44:1–67; Ex. 1027-15 ¶
`
`43; Ex. 1027-27–29, ¶¶ 73–76; Ex. 1027-43–45, ¶¶ 109–113.)
`
`46. Fourth, the hyperemia reported for Compound C is at most only
`
`modest. Notably, none of the animals tested with Compound C exhibited the most
`
`severe signs of hyperemia (i.e., none of the animals tested with Compound C were
`
`assigned a score of “4,”) even at the highest dose. (Ex. 1003 at 17.) In addition, as
`
`a physician, I note that modest hyperemia is not a concern and that we do not
`
`recommend that patients discontinue glaucoma medication even when conjunctival
`
`hyperemia is seen.
`
`21
`
`Micro Labs Exhibit 1032-22
`
`
`
`47. Fifth, the most important considerations to the POSA in 1996 were
`
`increased IOP-lowering potency (to treat more severe cases of elevated IOP) and a
`
`long duration of action (to enable 24 hour dosing). (See Ex. 1037.) A lead
`
`compound that exhibits both of these characteristics, such as Compound C of
`
`Klimko, would clearly stand out among the others. An excellent lead compound
`
`candidate such as Compound C would not be discounted just because it might
`
`cause some redness of the eye, especially considering its effectiveness at lowering
`
`IOP and its longer duration of action.
`
`D. Data showing longer-lasting efficacy would lead the POSA to
`select Compound C as a lead compound.
`48. Dr. Fechtner opines that “the data in Klimko does not demonstrate any
`
`benefit in IOP lowering activity for compound C, as compared to Compounds A,
`
`B, and D.” (Ex. 2002 at 14, ¶ 31.) I disagree. As I explained in my May 1