`Filed on behalf of: Eli Lilly and Company
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`______________________
`
`
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`______________________
`
`ELI LILLY AND COMPANY
`Petitioner
`v.
`TEVA PHARMACEUTICALS INTERNATIONAL GMBH
`Patent Owner
`______________________
`
`IPR2018-01427
`Patent No. 8,597,649
`______________________
`
`DECLARATION OF DR. ANDREW CHARLES, M.D.
`
`
`
`
`
`
`
`

`

`U.S. Patent No. 8,597,649
`
`
`2.
`
`2.
`
`TABLE OF CONTENTS
`I.
`INTRODUCTION .............................................................................. 5
`II. MATERIALS CONSIDERED ............................................................ 5
`III. DR. FERRARI INCORRECTLY CHARACTERIZES THE
`STATE OF THE ART AS OF 2005 .................................................... 6
`A. Dr. Ferrari’s Characterization of the Art Conflicts With How
`Migraine Drugs Were Generally Developed in 2005 ................. 6
`Dr. Ferrari Fails to Account for the Full Scope of the Art as
`of 2005...................................................................................... 8
`1.
`Researchers Were Pursuing Both Acute and
`Preventative Migraine Treatments .................................. 9
`By 2005, Researchers Developed Several Agents That
`Blocked the CGRP Pathway for Either Acute and/or
`Preventative Treatment ................................................. 11
`IV. A POSA WOULD HAVE EXPECTED ANTAGONIZING CGRP
`TO BE SAFE .................................................................................... 13
`A. Dr. Ferrari Primarily Relies on Older Studies That Do Not
`Reflect the Effects of Antagonizing Endogenous CGRP ......... 14
`Dr. Ferrari Ignores More Recent Studies That Alleviated or
`Eliminated His Purported Safety Concerns ............................. 17
`1.
`Blocking the CGRP pathway does not worsen
`ischemic episodes in animals ........................................ 17
`Blocking the CGRP pathway does not alter cerebral
`and systemic hemodynamics in humans ........................ 20
`3. Many researchers, including Dr. Ferrari, expected that
`blocking the CGRP pathway would be a safe and
`effective way to treat migraine ...................................... 25
`
`B.
`
`B.
`
`
`
`2
`
`

`

`U.S. Patent No. 8,597,649
`
`
`C.
`
`Dr. Ferrari Misinterprets Tan 1995, Which with Other
`Studies Indicates that Long-Term CGRP Antagonism Was
`Expected to Be Safe ................................................................ 26
`1.
`Dr. Ferrari exaggerates the effects seen in Tan 1995
`and Wong ..................................................................... 27
`Tjen-A-Looi and the knock-out mice studies do not
`support Dr. Ferrari’s positions ...................................... 28
`
`2.
`
`3.
`
`Dr. Ferrari fails to consider animal studies reporting
`no safety issues associated with long-term CGRP
`antagonism .................................................................... 30
`D. Any Caution by Wimalawansa Was Resolved as of 2005 ....... 33
`E.
`A POSA Would Have Reasonably Expected That a
`Humanized Anti-CGRP Antagonist Antibody Would Be
`Safe and Effective in Treating Migraine ................................. 34
`THE POTENTIAL RISK OF STROKE IN MIGRAINEURS
`WOULD NOT HAVE DETERRED THE DEVELOPMENT OF A
`HUMANIZED ANTI-CGRP ANTAGONIST ANTIBODY ............. 35
`A.
`The Absolute Risk of Ischemic Episodes in Migraine Patents
`Was Very Low ........................................................................ 35
`A Potential At-Risk Population Would Not Have Deterred
`Drug Development for the Vast Majority of Migraine
`Patients ................................................................................... 37
`VI. DR. FOORD’S RELIANCE ON SPARE RECEPTOR THEORY
`AND CROSS-BINDING ARE AT BEST HYPOTHETICAL AND
`UNSUPPORTED .............................................................................. 38
`A.
`The Clinical Evidence Contradicts Dr. Foord’s Hypothetical
`Spare Receptor Theory ........................................................... 38
`Dr. Foord’s CGRP Cross-Binding Opinions Are Equally
`Unsupported ........................................................................... 40
`Dr. Foord’s Discussion of Wimalawansa Is Incorrect and
`Incomplete .............................................................................. 42
`3
`
`V.
`
`B.
`
`B.
`
`C.
`
`
`
`

`

`U.S. Patent No. 8,597,649
`
`
`B.
`
`VII. DR. RAPOPORT INCORRECTLY CHARACTERIZES
`STATEMENTS FROM RESEARCHERS AS “PRAISE” OF
`“THE CLAIMED INVENTION” ..................................................... 43
`A. Dr. Rapoport’s Arguments Are Based on at Best Two
`Antibody Products, Ajovy® and Emgality® ............................. 44
`The Statements That Dr. Rapoport Identifies Include Praise
`for Antibodies That Do Not Fall Within the Scope of the
`Claims..................................................................................... 44
`Documents Relied Upon by Dr. Rapoport Further Support
`That a Humanized Anti-CGRP Antagonist Antibody Would
`Have Been Obvious ................................................................ 46
`VIII. DR. RAPOPORT HAS NOT ESTABLISHED ANY
`UNEXPECTED RESULTS .............................................................. 48
`IX. CONCLUSION ................................................................................ 52
`
`
`
`C.
`
`
`
`4
`
`

`

`U.S. Patent No. 8,597,649
`
`
`I.
`
`INTRODUCTION
`I, Andrew C. Charles, M.D., have been retained by Finnegan,
`1.
`
`Henderson, Farabow, Garrett & Dunner, LLP on behalf of Eli Lilly and Company
`
`(“Lilly”) as an independent expert in the field of neurology, including the treatment
`
`of migraine, with a research focus on calcitonin gene-related peptide (“CGRP”).
`
`2.
`
`I previously provided a declaration dated August 8, 2018 in support of
`
`Lilly’s Petition for Inter Partes Review of U.S. Patent No. 8,597,649 (“the ’649
`
`patent”). (Ex. 1002.) My background and qualifications are set forth in my
`
`previous declaration. I now provide my expert opinions on certain arguments
`
`made by Dr. Ferrari (Ex. 2141), Dr. Foord (Ex. 2054), and Dr. Rapoport (Ex.
`
`2165) in their declarations.
`
`3.
`
`I am being compensated for the time I spend on this matter, but my
`
`compensation is not contingent upon my opinions or the outcome of this or any
`
`other proceeding.
`
`II. MATERIALS CONSIDERED
`In addition to the materials I considered previously, I have
`4.
`
`additionally considered the materials cited in this declaration, which are listed in
`
`Appendix A. To the extent I am provided with additional documents or
`
`information, including any declarations in this proceeding, I may offer further
`
`opinions.
`
`
`
`5
`
`

`

`U.S. Patent No. 8,597,649
`
`III. DR. FERRARI INCORRECTLY CHARACTERIZES THE STATE OF
`THE ART AS OF 2005
`A. Dr. Ferrari’s Characterization of the Art Conflicts With How
`Migraine Drugs Were Generally Developed in 2005
`Although researchers were actively making anti-CGRP antagonist
`
`5.
`
`antibodies (see, e.g., Exs. 1021, 1022, 1033, 1055; Ex. 1002, ¶¶116-122) and the
`
`art taught their use for therapeutic purposes (see, e.g., Exs. 1026-1028, 1096; Ex.
`
`1002, ¶¶108-111) before November 2005, Dr. Ferrari asserts that a person of
`
`ordinary skill in the art (“POSA”) would not have been motivated to make a
`
`humanized anti-CGRP antagonist antibody because of potential safety concerns.
`
`(Ex. 2141, ¶12.) I do not agree with Dr. Ferrari’s opinion.
`
`6.
`
`Setting aside that by 2005, a POSA would not have expected anti-
`
`CGRP drugs to pose safety concerns that would discourage their development
`
`(infra §IV), Dr. Ferrari’s assertion that a POSA would not have been motivated to
`
`make a humanized anti-CGRP antagonist antibody until its safety was established
`
`is inconsistent with how researchers had been developing drugs, including those
`
`for migraine, at that time. As with drug development in any medical field, drug
`
`development for migraine necessarily involves acknowledgement of safety issues.
`
`It is unreasonable and unrealistic to expect that following target validation, all of
`
`the necessary safety information in humans would be addressed before initiating
`
`the development of new therapies for testing and use in various systems with the
`
`
`
`6
`
`

`

`U.S. Patent No. 8,597,649
`
`
`goal of eventual use in the clinical setting. As a result, a POSA would have been
`
`motivated to make a humanized anti-CGRP antagonist antibody much earlier in the
`
`drug development process than Dr. Ferrari’s opinions have indicated.
`
`7.
`
`The development of BIBN4096BS is illustrative and supportive. First
`
`BIBN4096BS was tested in both in vitro and in vivo studies that confirmed its
`
`ability to block the CGRP pathway. (See Ex. 1024; Ex. 1002, ¶40.) Shortly
`
`thereafter, BIBN4096BS was evaluated in healthy human subjects and migraine
`
`patients, confirming its safety and efficacy. (See Exs. 1025, 1042, 2019; Ex. 1002,
`
`¶¶40-43.) In parallel, researchers continued to evaluate it in animals to further
`
`understand its site and mechanism of action. (See, e.g., Exs. 1090, 1263, 1318; Ex.
`
`1002, ¶¶40, 62, 102.) These studies further motivated researchers to continue
`
`pursuing BIBN4096BS in clinical trials while developing other CGRP-pathway
`
`antagonists. (See, e.g., Ex. 1296, Abstract, 1305 (reporting Phase II trial of MK-
`
`0974, initiated in December 2005).)
`
`8.
`
`As I previously testified, based on the information available in 2005, a
`
`POSA would have targeted CGRP to treat migraine and other chronic diseases.
`
`(Ex. 1002, ¶¶35-47, 60-67, 107-114.) Multiple studies had demonstrated that
`
`antagonizing the CGRP pathway is safe and effective in humans. (See, e.g., Ex.
`
`1002, ¶¶40-43; infra §IV.) Information available in 2005 would have motivated a
`
`POSA to make a humanized anti-CGRP antagonist antibody, which would then
`
`
`
`7
`
`

`

`U.S. Patent No. 8,597,649
`
`
`allow a POSA to gather additional information from in vitro, animal, and human
`
`studies. (Ex. 1002, ¶¶107-123, 129-132.) Consequently, Dr. Ferrari’s conclusion
`
`that a POSA would not have been motivated to make a humanized anti-CGRP
`
`antagonist antibody until safety concerns were dispelled is not correct. (Ex. 2141,
`
`¶12.)
`
`B. Dr. Ferrari Fails to Account for the Full Scope of the Art as of
`2005
`Dr. Ferrari asserted that “therapeutic anti-CGRP antibodies were not
`
`9.
`
`even on the radar of investigators prior to November 2005” because clinicians and
`
`researchers in 2005 were focused on small molecule CGRP receptor antagonists
`
`with short half-lives for the treatment of acute migraine. (Ex. 2141, ¶¶12, 21-25.)
`
`As I discussed previously, the prior art was not so limited. (Ex. 1002, ¶¶48-67,
`
`108-113, 116-122.)
`
`10. For example, various agents developed by 2005 were being developed
`
`for both the acute and preventative treatment of migraine. An agent with a short
`
`half-life often prompted researchers to develop similar agents with longer half-
`
`lives or explore different treatment regimens, such as daily dosing, for chronic
`
`treatment. Similarly, and as I previously stated, researchers’ focus was not limited
`
`to CGRP receptor antagonists. (Ex. 1002, ¶¶48-67, 108-113, 116-122.) Several
`
`prior art studies demonstrated that researchers were investigating CGRP
`
`
`
`8
`
`

`

`U.S. Patent No. 8,597,649
`
`antagonism generally—either by targeting CGRP receptors or the ligand itself—for
`
`both acute and chronic treatments. (Id.)
`
`1.
`
`Researchers Were Pursuing Both Acute and Preventative
`Migraine Treatments
`11. By 2005, seven different triptans had been developed for clinical use.
`
`(See generally Ex. 1293; Ex. 1308.) Sumatriptan was the first triptan approved in
`
`the early 1990s for treating acute migraine. Despite sumatriptan’s effectiveness in
`
`the acute treatment of migraine, researchers soon recognized that sumatriptan did
`
`not consistently prevent the recurrence of migraine, which sometimes occurred
`
`within 24 hours after the initial dose. (Ex. 1281, S74.) Researchers specifically
`
`attributed sumatriptan’s inability to block migraine recurrence to its “short plasma
`
`half-life,” (id., S76) and suggested repeated dosing of sumatriptan to treat or
`
`prevent recurrence of attacks. (Id., S74, S76.)
`
`12. Despite direct vasoconstrictor properties of triptans and their potential
`
`cardiovascular and cerebrovascular effects (see Ex. 2141, ¶21; Ex. 1281, S74-75),
`
`researchers developed additional, triptans with longer half-lives. For example,
`
`frovatriptan had a relatively longer half-life of about 26 hours. (Ex. 1293, S125-
`
`26.) These longer-acting triptans were intended to reduce recurrence of migraine,
`
`but were also considered as potential preventive therapies. (Id.) For instance,
`
`frovatriptan and naratriptan were considered as short-term preventive therapies
`
`(daily dosing for about a week) for menstrual migraine. (Id., S127.) Naratriptan
`
`
`
`9
`
`

`

`was also administered daily up to 1 year as a preventative treatment for chronic
`
`migraine with no serious adverse events. (See Ex. 1294, Abstract; Ex. 1295,
`
`U.S. Patent No. 8,597,649
`
`
`Abstract.)
`
`13. Ergots are another example of migraine therapies, which were initially
`
`developed as acute treatments, but were then used as preventative migraine
`
`treatments by 2005. Like triptans, ergots were known to have direct and longer-
`
`lasting vasoconstrictor properties. (Ex. 1316, 14; Ex. 1290, 654.) In fact, in some
`
`cases the vasoconstrictive effects of ergots were understood to last much longer
`
`than what was expected based upon plasma concentrations. (Ex. 1316, 10-11.)
`
`Nevertheless, recognizing the benefits of longer-lasting effects, researchers
`
`recommended ergots for treating patients with very long attacks or frequent
`
`headache recurrence and in some cases, with four consecutive doses. (See id., 16,
`
`Table 3.) By 2005, ergots had been used as a daily preventive therapy for migraine
`
`in select patients. (See Ex. 1311, 557-58.)
`
`14. The established use of triptans and ergots as both acute and preventive
`
`treatments undermines Dr. Ferrari’s characterization that researchers in 2005 were
`
`focusing only on acute treatments for migraine. (Ex. 2141, ¶¶21-25.) Rather,
`
`several therapies were being used not only as acute treatments, but also to prevent
`
`migraine.
`
`
`
`10
`
`

`

`U.S. Patent No. 8,597,649
`
`
`2.
`
`By 2005, Researchers Developed Several Agents That
`Blocked the CGRP Pathway for Either Acute and/or
`Preventative Treatment
`15. As Dr. Ferrari correctly summarizes, multiple antagonists against the
`
`CGRP receptor were under development by November 2005, confirming that
`
`researchers understood CGRP was a key target for migraine therapy. (See Ex.
`
`2141, ¶¶22-24; Ex. 1002, ¶¶35-45.) In fact, before December 2005, Dr. Rapoport
`
`was involved in developing MK-0974, a CGRP receptor antagonist like
`
`BIBN4096BS. (Ex. 1296, Abstract, 1305.) It was understood that “CGRP
`
`receptor antagonists lack direct vasoconstrictor activity.” (Id., 1305.) Dr. Ferrari
`
`also participated in a clinical trial evaluating MK-0974, echoing Dr. Rapoport’s
`
`belief that “antagonists of CGRP receptor do not seem to have direct
`
`vasoconstrictor properties.” (Ex. 1313, 2115.)
`
`16. While initially developed as acute treatments, researchers sought to
`
`extend blocking the CGRP pathway to preventative treatments. The prior art
`
`development of CGRP8-37 and its conjugates is illustrative. CGRP8-37 is a peptide
`
`CGRP receptor blocker. (Ex. 1002, ¶¶20, 48.) Although CGRP8-37 antagonizes
`
`CGRP in vitro and in vivo, as Dr. Ferrari noted, CGRP8-37 was not pursued as a
`
`therapeutic due to its short half-life. (Ex. 2141, ¶23; Ex. 1002, ¶90.) But
`
`researchers continued to develop related compounds with longer half-lives, aiming
`
`to use them as acute and preventive treatments. Such efforts led to the
`
`
`
`11
`
`

`

`U.S. Patent No. 8,597,649
`
`
`development of conjugated CGRP fragments with significantly increased half-
`
`lives. (See Ex. 2006, 7:1-7 (discussing the desirability of conjugated CGRP
`
`peptides providing increased pharmacological half-life and decreased
`
`immunogenicity); id., 7:35-41 (“the CGRP peptide antagonists [providing]
`
`substantially increased in vivo pharmacological half-life”).) Indeed, one such
`
`fragment showed a half-life of more than 40 hours in rats, offering the possibility
`
`for both acute and preventive treatments. (Id., col. 121 (Table 10); id., 8:12-17
`
`(“The present invention is also directed to a method of preventing or mitigating
`
`migraine, which involves administering . . . prophylactically effective amount of
`
`the inventive composition of matter[.]”).)
`
`17. Various antagonists directed to CGRP itself were also under
`
`development in 2005. As I previously testified, researchers had already made
`
`several anti-CGRP antagonist antibodies (see, e.g., Exs. 1021, 1022, 1033, 1055;
`
`Ex. 1002, ¶¶116-122.) Researchers were also developing aptamers that bound to
`
`and blocked CGRP from binding to its receptor for the treatment of migraine,
`
`recognizing that such aptamers can be stabilized to achieve longer half-lives. (See,
`
`e.g., Ex. 1082, Abstract (“Stabilized aptamers and Spiegelmers have shown
`
`activity in vivo and may be used as therapeutics.”), 2; Ex. 1240, 923 (evaluating “a
`
`bio-stable aptamer” binding to CGRP and recommending it as “a new therapeutical
`
`strategy in diseases . . . such as migraine and other primary headaches”).) Indeed,
`
`
`
`12
`
`

`

`U.S. Patent No. 8,597,649
`
`
`they were known as “nucleic analogs to antibodies” due to their specificity,
`
`biological activity, favorable safety profile, and potential long in vivo half-lives
`
`ranging from hours to days. (Ex. 1309, Abstract.)
`
`IV. A POSA WOULD HAVE EXPECTED ANTAGONIZING CGRP TO
`BE SAFE
`18. Dr. Ferrari asserts that a POSA would have been dissuaded from
`
`making a humanized anti-CGRP antagonist antibody because of potentially life-
`
`threatening side effects. (Ex. 2141, ¶28.) Dr. Ferrari identifies two potential side
`
`effects of long-term CGRP antagonism. (Id., ¶37.) First, Dr. Ferrari asserts that a
`
`POSA would have expected inhibition of CGRP to result in a worsening of life-
`
`threatening ischemic episodes such as stroke and heart attacks (myocardial
`
`infraction, “MI”). (Id., ¶¶37-47.) Second, Dr. Ferrari asserts that a POSA would
`
`have expected inhibition of CGRP to increase blood pressure. (Id., ¶¶37, 48-54.) I
`
`disagree with Dr. Ferrari’s conclusions.
`
`19. By November 2005, a POSA would have understood that the
`
`purported safety concerns identified by Dr. Ferrari were theoretical (at best) and
`
`contradicted by more recent animal and human studies. As explained in more
`
`detail below, Dr. Ferrari primarily relies on outdated studies that did not reflect the
`
`consequences of antagonizing naturally-present CGRP, i.e., endogenous CGRP.
`
`By 2005, these older studies had been superseded by numerous animal and clinical
`
`studies demonstrating that blocking the endogenous CGRP pathway does not
`
`
`
`13
`
`

`

`U.S. Patent No. 8,597,649
`
`
`increase blood pressure and does not worsen ischemic episodes. (E.g., Exs. 1283,
`
`1284, 1285, 1318, 1263, 1240, 1025, 1042, 2019.) Based on these subsequent
`
`studies, leading figures in the field of migraine expected that CGRP antagonism
`
`would be a safe and effective way to treat migraine. Further, the outdated safety
`
`concerns certainly did not stop those in the art from pursuing anti-CGRP drugs for
`
`the treatment of migraine, regardless of their duration of action. Indeed, therapies
`
`with longer duration of action were being actively pursued. (Supra §III.B.)
`
`A. Dr. Ferrari Primarily Relies on Older Studies That Do Not
`Reflect the Effects of Antagonizing Endogenous CGRP
`20. Dr. Ferrari asserts that a POSA would have expected a worsening of
`
`ischemic episodes because CGRP was purportedly known to have cardio- and
`
`cerebral protective effects. (Ex. 2141, ¶¶38-47.) To support this assertion, Dr.
`
`Ferrari relies on a group of experimental or observational studies, all of which
`
`were published between 1990 and 1998. (Id. (citing Exs. 2058, 2070, 2079, 2089,
`
`2139, 2150, 2151, 2152, 2154, and 2209).)
`
`21. These studies, however, reveal little about the role of endogenous
`
`CGRP, failing to show that endogenous CGRP possesses cardio- and cerebral
`
`protective effects. Uren (Ex. 2058), for example, administered exogenous CGRP
`
`to patients (nine males, three females) with angina. (Ex. 2058, Abstract, 1477; Ex.
`
`2141, ¶¶38, 44.) Antagonism of naturally present CGRP was not evaluated in this
`
`study. (Ex. 2058, Abstract, 1477.) Likewise, Juul (Ex. 2139) administered
`
`
`
`14
`
`

`

`U.S. Patent No. 8,597,649
`
`
`exogenous CGRP to five patients (three males, two females) with subarachnoid
`
`hemorrhage. (Ex. 2139, Abstract, Table 1; Ex. 2141, ¶¶40, 45.) Gennari (Ex.
`
`2079) administered exogenous βCGRP to a small number of patients (5) with
`
`congestive heart failure. (Ex. 2079, Abstract, 239; Ex. 2141, ¶39.) All of these
`
`studies increased CGRP levels in patients with preexisting ischemic conditions as
`
`opposed to antagonizing endogenous CGRP.
`
`22. Other studies relied on by Dr. Ferrari (Exs. 2070, 2089, 2151, and
`
`2154) were simply observational, i.e., they monitored CGRP levels with no
`
`intervention whatsoever. (See Ex. 2070, Abstract (reporting CGRP level changes
`
`during the day in humans); Ex. 2089, Abstract (reporting CGRP levels in rat
`
`tissues and plasma); Ex. 2151, 168 (reporting CGRP levels in patients following
`
`acute myocardial infarction, several medicated with urokinase); Ex. 2154, Abstract
`
`(reporting CGRP levels in cerebral vessels isolated from deceased patients
`
`following subarachnoid hemorrhage); Ex. 2141, ¶¶38, 42, 45.) These studies also
`
`provide no information regarding the consequences of antagonizing endogenous
`
`CGRP.
`
`23. Dr. Ferrari also relies on studies that indiscriminately blocked CGRP
`
`and other neuropeptides, making it impossible to determine whether any of the
`
`observed effects resulted from the presence or absence of CGRP. For instance,
`
`Edvinsson (Ex. 2209) created a surgical lesion of the trigeminal nerve, which
`
`
`
`15
`
`

`

`U.S. Patent No. 8,597,649
`
`
`eliminated CGRP and another neuropeptide (substance P), and likely had other
`
`effects. (See Ex. 2209, Abstract; Ex. 2141, ¶40.) Edvinsson (Ex. 2209) also
`
`describes in vitro studies with exogenously added CGRP or capsaicin, the latter of
`
`which broadly depletes CGRP and other stored neuropeptides. (Ex. 2209,
`
`Abstract, 376-77.) The same applies to Källner (Ex. 2150), where capsaicin was
`
`used to broadly deplete CGRP and other neuropeptides. (See Ex. 2150, Abstract;
`
`Ex. 2141, ¶38.) As a result, one cannot ascribe the observed results to any one
`
`peptide in particular.
`
`24. Notably, Dr. Ferrari relies upon Li (Ex. 2152), but fails to indicate that
`
`in the absence of exogenous CGRP, CGRP8-37 showed no effect on cardiac
`
`function. (Ex. 2152, 165; Ex. 2141, ¶38.) In other words, in that study, consistent
`
`with other studies that I describe below, blocking the endogenous CGRP pathway
`
`had no observed cardiovascular effect.
`
`25. Collectively, the studies that Dr. Ferrari relies upon do not support his
`
`contention that a POSA would have had concerns that antagonizing endogenous
`
`CGRP would worsen ischemic episodes.
`
`
`
`16
`
`

`

`U.S. Patent No. 8,597,649
`
`
`B. Dr. Ferrari Ignores More Recent Studies That Alleviated or
`Eliminated His Purported Safety Concerns
`Blocking the CGRP pathway does not worsen ischemic
`1.
`episodes in animals
`26. While relying on studies published well before 2000, Dr. Ferrari
`
`ignores other more recent, prior-art animal studies demonstrating that blocking the
`
`CGRP pathway does not worsen ischemic episodes.
`
`27. A study from Källner and colleagues—that has authors that overlap
`
`with the authors of Ex. 2150—demonstrated that antagonizing endogenous CGRP
`
`using a specific CGRP pathway antagonist, CGRP8-37, does not worsen myocardial
`
`infarction in pigs. (Ex. 1283, 494 (evaluating ischemic pigs of either sex); id., 498
`
`(“This study showed no effect of CGRP or CGRP(8-37) on infarct size.”).) The
`
`absence of any effect led to the conclusion that “locally released CGRP does not
`
`function as a cardioprotective agent in this experimental model.” (Id., 498.)
`
`28. A study published in 2001 also showed that endogenous CGRP did
`
`not affect myocardial infarcts. (Ex. 1284, Abstract.) In this study, Wu and
`
`colleagues antagonized endogenous CGRP using BIBN4096BS in myocardially
`
`ischemic rats and confirmed that CGRP antagonism showed no effect on
`
`myocardial infarct size. (Id., 591-92; see also id., Abstract (“the CGRP antagonist
`
`BIBN4096BS before occlusion until the end of reperfusion had no statistically
`
`significant effect on the infarct size”).) Moreover, consistent with studies
`
`
`
`17
`
`

`

`U.S. Patent No. 8,597,649
`
`
`administering exogenous CGRP that Dr. Ferrari relies upon (see, e.g., Exs. 2058,
`
`2079, 2139), CGRP’s cardioprotective effect was observed only when exogenous
`
`CGRP (about 10-fold excess over endogenous CGRP) was administered,
`
`suggesting that “[o]nly high plasma levels [of] CGRP may cause cardioprotection.”
`
`(Ex. 1284, 593.)
`
`29. A study published in 2003 further demonstrated that blocking the
`
`CGRP pathway has no effect on cardiac function during the late stage of heart
`
`failure. (Ex. 1285, Abstract.) In this study, blocking the CGRP pathway using
`
`CGRP8-37—at a dose achieving complete blockade—caused no significant changes
`
`in cardiac contractility and peripheral vascular dynamics. (Id., 656 (evaluating
`
`dogs of either sex); id., 658 (“[The LV dP/dt, mean arterial pressure, mean left
`
`arterial pressure, and heart rate] were unchanged at the heart failure stage during
`
`intravenous infusion of α-CGRP[8-37] (30 µg/kg/min). No significant changes
`
`were observed (P > 0.05) in regional blood flow in myocardium, kidney, brain,
`
`spleen, stomach, and large and small intestine, muscle and fat during the infusion
`
`of α-CGRP[8-37] … .”).) The authors concluded that “endogenous α-CGRP does
`
`not appear to play a major role in regulation of the cardiovascular system in
`
`advanced heart disease.” (Id., 660.)
`
`30. Dr. Ferrari also did not consider more recent studies evaluating the
`
`role of endogenous CGRP in regulating systemic and regional hemodynamics.
`
`
`
`18
`
`

`

`U.S. Patent No. 8,597,649
`
`
`(Ex. 1318, Abstract; Ex. 1263, Abstract.) In a 2003 study, repeated administration
`
`of BIBN4096BS to anesthetized pigs did not affect heart rate, mean arterial blood
`
`pressure, and systemic vascular conductance, suggesting that endogenous CGRP
`
`“does not seem to play an important physiological role in regulating basal vascular
`
`tone.” (Ex. 1318, Abstract.) Using anesthetized rats, a 2004 study again
`
`demonstrated that administering BIBN4096BS, which binds to rat CGRP receptors
`
`with a binding affinity within a nanomolar range (Ex. 1024, 422), did not alter
`
`regional and systemic hemodynamic variables including heart rate, mean arterial
`
`blood pressure, cardiac output, systemic vascular conductance, and regional
`
`vascular conductance in numerous tissues including brain and heart. (Ex. 1263,
`
`293.) Consistent with the 2003 study, the authors of this study concluded that
`
`“endogenous CGRP does not play an important role in regulating systemic and
`
`regional heamodynamics.” (Id., 296.)
`
`31.
`
`In a 2005 study that Dr. Ferrari did not consider, a CGRP-binding
`
`aptamer (i.e., an agent that binds CGRP directly as opposed to its receptor) was
`
`systemically administered to rats and shown to inhibit increases in meningeal
`
`blood flow caused by electrical stimulation. (Ex. 1240, 923.) Notably, the study
`
`reported that “[b]asal blood flow and systemic arterial pressure were unchanged”
`
`by this CGRP-targeting aptamer. (Id.) This study further demonstrates that a
`
`POSA would have expected that blocking the CGRP pathway, either by targeting
`
`
`
`19
`
`

`

`U.S. Patent No. 8,597,649
`
`
`CGRP receptor or CGRP ligand itself, would not significantly affect vascular
`
`functions mediated by endogenous CGRP.
`
`32. These additional animal studies, all published before November 2005,
`
`strongly indicated that blocking endogenous CGRP does not worsen ischemic
`
`episodes. These studies also undermine Dr. Ferrari’s assertion that endogenous
`
`CGRP plays an important role in regulating systemic and regional hemodynamics
`
`“in a patient’s baseline state.” (See Ex. 2141, ¶48.) As summarized below, several
`
`clinical studies in human subjects further established that blocking the CGRP
`
`pathway did not raise significant safety concerns.
`
`2.
`
`Blocking the CGRP pathway does not alter cerebral and
`systemic hemodynamics in humans
`33. By November 2005, several clinical studies in humans had
`
`demonstrated that blocking the CGRP pathway does not alter cerebral and systemic
`
`hemodynamics in a patient’s baseline state. The Phase II trial with BIBN4096BS,
`
`discussed at length in my earlier declaration, is one example. (See Ex. 1025; Ex.
`
`1002, ¶¶40-43.) Olesen et al. evaluated several outcome measures, including the
`
`frequency of adverse events, changes from base line in heart rate, blood pressure
`
`on supine position, and clinical laboratory values and reported that paresthesia was
`
`the only adverse event of note. (Ex. 1025, 1106-07, 1109.) The favorable safety
`
`and tolerability profile reported in earlier Phase I studies was confirmed. (Id.,
`
`1109.) Because BIBN4096BS had not shown any vasoconstrictor activity in
`
`
`
`20
`
`

`

`U.S. Patent No. 8,597,649
`
`
`animals and humans, Olesen concluded that “BIBN 4096 BS does not seem to
`
`have vasoconstrictor properties” and, if subsequent studies confirm the same, “this
`
`will represent an advantage over the triptans.” (Id.)
`
`34. Despite Olesen’s express recognition of BIBN4096BS’s favorable
`
`safety profile, Dr. Ferrari asserts that little can be gleaned from it because Olesen’s
`
`data base was purportedly too small. (Ex. 2141, ¶22.) A POSA, however, would
`
`have considered this study as part of the growing body of work (e.g., additional
`
`animal and clinical studies), establishing that the CGRP-pathway could be
`
`antagonized without the vasoconstrictive properties of triptans. As a result, a
`
`POSA would have viewed Olesen’s study—and his comments about
`
`BIBN4096BS’s lack of vasoconstrictive effects—as a further indication that
`
`blocking the CGRP pathway was expected to be both safe and effective in humans.
`
`35. For example, Iovino and colleagues evaluated BIBN4096BS in
`
`healthy individuals and confirmed that blocking the CGRP pathway did not alter
`
`systemic hemodynamics including blood pressure. (Ex. 1042, Abstract; Ex. 1002,
`
`Appendix B.) Indeed, the authors observed no clinically relevant changes in blood
`
`pressure, pulse rate, respiratory rate, ECG, laboratory tests, and peripheral blood
`
`flow. (Ex. 1042, Abstract.) The apparent lack of vasoconstrictive effects led the
`
`authors to state that “CGRP antagonists have the potential to be a class of
`
`antimigraine drugs without cardiovascular side effects.” (Id., 654.)
`
`
`
`21
`
`

`

`U.S. Patent No. 8,597,649
`
`36. A study from Petersen and colleagues again confirmed that circulating
`
`CGRP is “not likely to exert a vasodilatory activity in the resting state.” (Ex. 2019,
`
`Abstract.) In the paper, the authors hypothesized that “[i]f CGRP plays a role in
`
`the regulation of the vascular tone of cerebral arteries or [cerebral blood flow] in
`
`human volunteers, the administration of a potent CGRP-antagonist could be
`
`expected to induce vasoconstriction of cerebral arteries and/or arterioles.” (Id.,
`
`145.) To evaluate this hypothesis, the authors administered BIBN4096BS to
`
`heathy individuals and measured both systemic and cerebral hemodynamic values.
`
`(Id., 142-43.) Consistent with Iovino’s study, the authors in this study found no
`
`significant changes in systemic hemodynamic values, including systolic blood
`
`pressure, diastolic blood pressure, and heart rate. (Id., 143.) Likewise, the authors
`
`reported no significant changes in cerebral blood flow or the diameter of the
`
`middle cerebral artery. (Id., 142-43.) From these results, the authors concluded
`
`that “[CGRP] has no influence on the resting vascular tone of cerebral arteries,”
`
`further supporting “cerebrovascular safety of CGRP-antagonism in the treatment of
`
`acute migraine.” (Id., 14