`Wockhardt Bio AG, Teva Pharmaceuticals USA, Inc.,
`Aurobindo Pharma U.S.A. Inc., and Sun Pharmaceutical
`Industries, Ltd., Sun Pharma Global FZE and
`Amneal Pharmaceuticals LLC,
`Petitioners
`v.
`AstraZeneca AB,
`Patent Owner
`
`IPR2015-01340
`US RE44,186 E
`
`January 25, 2017
`
`
`
`Otsuka v. Sandoz, 678 F.3d 1280, 1292 (Fed. Cir. 2012)
`
`“In determining whether a chemist would
`have selected a prior art compound as a
`lead, the analysis is guided by evidence of
`the compound’s pertinent properties.”
`
`Source: Institution Dec. at 9.
`
`2
`
`
`
`Daiichi Sankyo v. Matrix Labs., 616 F.3d 1346, 1354 (Fed. Cir. 2010)
`
`in
`“[T]he lead compound analysis must,
`keeping with KSR, not rigidly focus on the
`selection of a single, best lead compound . . .”
`
`Source: Institution Dec. at 10.
`
`3
`
`
`
`Ashworth 25 is a Pertinent Lead Compound
`
`P2
`Substituent
`
`P1
`Substituent
`
`Ashworth 25
`
` Potency:
`Ki < 2 nM
` Solution Stability: t1/2 > 48 hours
`
`Source: EX1007 (Ashworth I) at 1166.
`
`4
`
`
`
`Ashworth 25 Has Good Potency and Stability
`
`Source: EX1007 (Ashworth I) at 1165‐66.
`
`5
`
`
`
`Measuring Potency by Ki and IC50 Values
`
`Smaller Ki and IC50 values
`represent greater potency
`
`the “concentration of
`is
`IC50
`inhibitor producing 50 per cent
`inhibition[.]”
`
`Ki is the “dissociation constant of the enzyme‐
`inhibitor complex, or the reciprocal of the
`binding affinity of
`the inhibitor
`to the
`enzyme[.]”
`
`Source: EX1025 (Cheng) at 3099; EX1003 (Rotella Decl.), ¶64.
`
`6
`
`
`
`Ashworth I Table II Compounds
`
`Source: EX1007 (Ashworth I) at 1166.
`
`7
`
`
`
`Cyanopyrrolidines Were of Interest
`
`NVP‐DPP728, containing a
`cyanopyrrolidine, entered clinical
`trials in humans.
`
`NVP‐DPP728
`“By the time of the invention,
`Novartis had selected NVP‐DPP728
`as a clinical candidate, and it was
`reported to be safe and effective
`in initial studies in humans.”
`Source: PO Resp. at 12, 14; EX2012 (Rothenberg) at A39; EX1074 (Second Rotella Decl.), ¶15.
`
`8
`
`
`
`Ashworth 25 More Potent than Other Clinical Candidates
`
`NVP‐DPP728
`
`Ki = 11 nM
`
`P32/98
`
`IC50 = 2800 nM
`
`Ashworth 25
`
`Ki = 1.4 nM
`
`Dr. Weber concedes that “compound 25 of Ashworth‐I
`[is] more potent than the clinical candidates NVP‐
`DPP728 and P32/98[.]”
`
`Source: EX1074 (Second Rotella Decl.), ¶11; EX2016 (Hughes) at 11600;
`EX1007 (Ashworth I) at Table 2, 1166; EX2078 (Schon) at 308; EX2056 (Weber Decl.), ¶172 .
`
`9
`
`
`
`Superior Stability of Ashworth 25
`
`Ashworth II compound 3
`
`Ashworth I compound 25
`
`t1/2 = 27 h
`
`t1/2 = >48 h
`
`Ashworth 25 has greater in vitro stability (longer half‐
`life, t1/2) than compound 3 from Ashworth II.
`
`Source: EX1074 (Second Rotella Decl.), ¶14; EX2001 (Ashworth II) at 2748.
`
`10
`
`
`
`Summary of Structural Differences
`
`Ashworth 25
`
`Cyclopropanation of the
`pyrrolidine ring
`o EX1007: Ashworth I
`o EX1010: Hanessian
`
`c
`Replace cyclohexyl ring with
`hydroxyadamantyl
`o EX1007: Ashworth I
`o EX1008: Villhauer WO 98
`o EX1009: Raag
`
`Saxagliptin
`
`11
`
`
`
`Petitioner’s Motivation to Modify P1
`
`Cyclopropanation
`o flattens and rigidifies
`pyrrolidine ring
`o modulates cyano position
`Optimizes interaction with
`DPP‐4 enzyme to improve
`activity and stability
`
`P1
`Substituent
`
`Source: Pet. at 29; EX1003 (First Rotella Decl.), ¶134; EX1010 (Hanessian).
`
`12
`
`
`
`Cyclopropanation Modulates Proline Conformation
`
`“The structures and conformations of 6 and 8 in the
`solid state were unambiguously confirmed by single‐
`crystal X‐ray analysis. Table 1 lists selected torsion
`angles for compounds 6 and 8, where considerable
`‘flattening’ of the pyrrolidine ring is observed relative
`to N‐Boc‐L‐proline[.]”
`
`CO2H
`
`N
`Boc
`N-Boc-
`L-proline
`
`CO2H
`
`N
`Boc
`6
`
`CO2H
`
`N
`Boc
`8
`
`Source: EX1010 (Hanessian ‘97) at 1882; EX1003 (Rotella Decl.), ¶135.
`
`13
`
`
`
`Cyclopropanation Confers Conformational Rigidity
`
`Source: EX2174 (Rotella Depo. Trans.), 119:10‐21.
`
`14
`
`
`
`Increase in Cyanopyrrolidine Stability Was Predictable
`
`Source: EX2174 (Rotella Depo. Trans.), 139:22‐140:3; EX2002 (Magnin) at 2591.
`
`15
`
`
`
`Patent Owner Cites Augustyns 1997
`
`“[T]he inhibitory capacity increases from
`azetidine 8b (270 μM) and pyrroline 9b (100
`μM) to the optimal five‐rings pyrrolidine 3
`(21 μM) and thiazolidine 4 (18 μM), and
`decreases
`when
`larger
`rings
`like
`tetrahydropyridine 10b (310 μM), piperidine
`6b (510 μM) or hexamethyleneimine 7b
`(2700 μM) are at the P1 position.”
`
`Patent Owner: “Increasing the
`pyrrolidine ring size to a 6‐ or
`7‐membered ring . . . was not
`well tolerated.”
`
`Source: Patent Owner Resp. at 10; EX2151 (Augustyns 1997) at 303.
`
`16
`
`
`
`Small Changes to P1 Ring Size Were Tolerated
`
`Ashworth II compound 5
`
`Ashworth II compound 3
`
`Ki = 2.2 nM
`
`Ki = 0.41 nM
`
`Source: EX2001 (Ashworth II) at 2747; EX1074 (Second Rotella Decl.), ¶40.
`
`17
`
`
`
`Cyclopropanation Has Minimal Effect on Ring Size
`
`“[T]hese compounds have a cyclopropane ring in place
`of the choline moiety and are considered to be the
`smallest chemical structure among Ach derivatives
`capable of conferring conformational rigidity.”
`
`Source: EX1021 (Chiou) at 243; EX1003 (Rotella Decl.), ¶135; Pet. at 22.
`
`18
`
`
`
`Cyclopropanation Fits with 5‐Membered Ring Preference
`
`A POSA “would understand that cyclopropanation
`preserves the saturated five‐membered ring, while also
`providing controlled modifications to the pyrrolidine
`conformation, as taught by Hanessian.”
`
`“Therefore, we believe that the S‐1
`subsite of DPP IV ideally fits a five‐
`membered saturated ring.”
`
`Source: EX2151 (Augustyns) at 303; EX1074 (Second Rotella Decl.), ¶40.
`
`19
`
`
`
`Methods For Cyclopropanating Pyrrolidine were Known
`
`Hanessian teaches
`synthesis of cis‐4,5 and
`trans‐4,5 cyclopropyl
`cyanopyrrolidine.
`Source: EX1010 (Hanessian ‘97) at 1882; EX1074 (Second Rotella Decl.), ¶38.
`
`4
`
`5
`
`20
`
`
`
`Limited Number of Positions to Cyclopropanate Pyrrolidine
`
`Only 5 possible ways to
`cyclopropanate pyrrolidine ring
`(including cis and trans isomers).
`
`Source: EX1010 (Hanessian ‘97) at 1882; EX1003 (Rotella Decl.), ¶¶139‐40.
`
`21
`
`
`
`Magnin Confirms Ease of Evaluating Each Option
`
`Magnin confirms the straightforward task of screening
`cyclopropanation derivatives at each of the available
`cyanopyrrolidine positions.
`Source: EX2002 (Magnin) at 2588; EX1074 (Second Rotella Decl.), ¶50.
`
`22
`
`
`
`Motivation to Modify P2
`
`Sterically bulky P2 substituents
`o improve stability
`o improve potency
`o favor trans confirmation
`and reduce cyclization
`
`P2
`Substituent
`
`Cyclization
`
`Source: Pet. at 7, 24; EX1003 (Rotella Decl.), ¶¶ 114‐15; EX1007 at 1165‐66.
`
`23
`
`
`
`Intramolecular Cyclization Favored in Cis‐Conformation
`
`Intramolecular cyclization is minimized by favoring trans‐
`conformation, instead of cis‐conformation.
`
`Source: EX1003 (Rotella Decl.), ¶¶46‐47; EX1015 (Lin) at 14022; Pet. at 7.
`
`24
`
`
`
`Bulky P2 Groups Improve Stability
`
`Source: EX1007 (Ashworth I) at 1165‐66; EX1074 (Second Rotella Decl.), ¶19.
`
`25
`
`
`
`Steric Bulk Localized at the Beta Position
`
`Beta
`position
`
`Beta
`position
`
`Ashworth I Compound 25
`t1/2= >48 h
`Ki = 1.4 ± 0.5 nm
`
`Ashworth I Compound 27
`t1/2= >48 h
`Ki = 3.8 ± 0.8 nm
`
`Source: EX1007 (Ashworth I) at Table 2; EX1074 (Second Rotella Decl.), ¶24.
`
`26
`
`
`
`Steric Bulk Localized at the Beta Position
`
`Beta
`position
`
`Beta
`position
`
`Ashworth I Compound 25
`t1/2= >48 h
`Ki = 1.4 ± 0.5 nm
`
`Ashworth I Compound 24
`t1/2= 48 h
`Ki = 1.1 ± 0.2 nm
`
`Source: EX1007 (Ashworth I) at Table 2; EX1074 (Second Rotella Decl.), ¶24.
`
`27
`
`
`
`More Atoms Alone ≠ Steric Bulk
`
`Lys(Z) group is longer with
`more atoms.
`
`Lys(Z) is less sterically bulky
`compared to a cyclohexyl
`group, with a smaller
`footprint at the β‐position.
`
`Lys(Z) has decreased
`stability compared to a
`cyclohexyl substituent.
`
`Lys(Z)/
`Lys(Cbz)
`
`Beta
`position
`
`Ashworth I Compound 28
`t1/2= 24 h
`
`Source: EX1007 (Ashworth I) at Table 2; EX1074 (Second Rotella Decl.), ¶24; EX2261 (PO Demonstrative).
`
`28
`
`
`
`Use of Adamantyl at P2 was Known in DPP‐4 Inhibitors
`
`Source: EX1008 (Villhauer WO ’98) at 13; EX1074 (Second Rotella Decl.), ¶20.
`
`29
`
`
`
`Hydroxyadamantyl Was Known From Villhauer 2000
`
`Source: EX1073 (Weber Depo. Trans.), 63:2‐8; EX1074 (Second
`Rotella Decl.), ¶33; EX2013 (Villhauer 2000), 7:15‐27 .
`
`30
`
`
`
`Hydroxyadamantyl at P2 was Known in DPP‐4 Inhibitors
`
`Vildagliptin
`
`Source: EX2013 (Villhauer 2000), 7:15‐27; Pet. Reply at 14.
`
`31
`
`
`
`Patent Owner Cites Mentlein 1993
`
`“Preferential amino acids for the
`P’1 position are not known.”
`
`Source: EX2056 (Weber Decl.), ¶91; EX2096 (Mentlein) at 833.
`
`32
`
`
`
`Mentlein: Bulky Groups are Preferred at P2
`
`“In the P2 position bulky amino acids with
`an obligate free amino group are preferred.”
`
`Source: EX2096 (Mentlein) at 833; EX1074 (Second Rotella Decl.), ¶19.
`
`33
`
`
`
`Hydroxyadamantyl is a Bulky Group
`
`Source: Pet. Reply at 13; EX1073 (Weber Depo. Trans.), 60:11‐19.
`
`34
`
`
`
`Metabolites Guide Modification of Drug Candidates
`
`“Metabolite identification in drug discovery provides
`early information that can lead to structural changes
`in the current
`lead compound,
`improving such
`pharmacokinetic parameters as oral bioavailability,
`half‐life (t1/2), or Cmax.”
`
`Source: EX1020 (Korfmacher) at 534; EX1003 (Rotella Decl.), ¶53.
`
`35
`
`
`
`Metabolites Guide Modification of Drug Candidates
`
`provide
`can
`identification
`“Early metabolite
`information on how to improve the metabolic stability
`of
`the lead structure.
`In this way,
`future lead
`compounds might be a metabolite identified from the
`previous lead drug or an analogue of the previous
`drug designed to block
`the major
`route of
`metabolism.”
`
`Source: EX1020 (Korfmacher) at 534; EX1003 (Rotella Decl.), ¶53.
`
`36
`
`
`
`Adamantane is Metabolized at Tertiary Carbons
`
`substrate we have
`the only
`“Adamantane is
`investigated, in this study, that is metabolized to a
`single product despite having a relatively high active
`site mobility. The single product can be attributed to
`the existence of only two types of unique carbon
`atoms in adamantane, together with the greater
`reactivity of tertiary versus secondary carbons.”
`
`Source: EX1009 (Raag) at 2678; EX1003 (Rotella Decl.), ¶¶ 80, 126.
`
`37
`
`
`
`Hydroxylated Adamantane Can Impede Metabolism
`
`“Blocking metabolism at the 3‐position would result in greater
`metabolic stability.”
`
`Source: EX1003 (Rotella Decl.), ¶¶ 53, 127; EX1020 (Korfmacher) at 534.
`
`38
`
`
`
`Hydroxyadamantyl at P2 was Known in DPP‐4 Inhibitors
`
`Tertiary
`carbon
`
`Vildagliptin
`
`Source: EX2013 (Villhauer 2000), 7:15‐27; Pet. Reply at 14.
`
`39
`
`
`
`Saxagliptin is also Metabolized at a 3˚ Carbon
`
`3˚
`
`3˚
`
`3˚
`
`Saxagliptin
`
`Saxagliptin Metabolite M2
`
`M2 metabolite predictably results from a second oxidation at one
`of only two remaining tertiary adamantyl carbons.
`
`Source: EX2045 (Su) at 1346; EX1074 (Second Rotella Decl.), ¶32.
`
`40
`
`
`
`Summary of Structural Modifications
`
`Cyclopropanate the pyrrolidine
`ring
`o Smallest possible fusion confers
`conformational rigidity and modulate
`position of cyano group
`Source: EX1007 (Ashworth I) at 1163
`EX1010 (Hanessian) at 1882
`Replace cyclohexyl ring with
`hydroxyadamantyl
`c
`o Sterically bulky substituent favors trans
`conformation to maintain or improve
`stability and prevent intramolecular
`cyclization; potential to improve potency
`Source: EX1007 (Ashworth I) at 1163
`EX1008 (Villhauer) at 13
`o Routine evaluation of metabolites
`Source: EX1009 (Raag)
`
`Ashworth 25
`
`Saxagliptin
`
`41
`
`
`
`Secondary Considerations Don’t Overcome Prima Facie Case
`
`Did not meet a long‐felt need
`
`No commercial success
`
`No evidence of failure of others
`
`No unexpected results
`
`42
`
`
`
`Metformin is Still the Most Preferred Anti‐Diabetic Drug
`
`“Metformin has the strongest
`evidence base and demon‐
`strated long‐term safety as
`pharmacological
`therapy for
`diabetes prevention.”
`
`“In my opinion, there is no drug
`(including saxagliptin) that stands
`out above the others as the
`c
`second best choice alternative to
`metformin.”
`
`Source: EX2065 (American Diabetes) at 44; EX1041 (Tanenberg Decl.), ¶¶20‐21; Pet. Reply at 4.
`
`43
`
`
`
`Metformin is Essential Medicine for Type 2 Diabetes
`
`The World Health Organization DOES NOT
`list saxagliptin as an essential medicine in
`the treatment of Type 2 Diabetes.
`
`Source: EX1043 (WHO) at 34; EX1041 (Tanenberg Decl.), ¶¶19, 21.
`
`44
`
`
`
`Metformin is Still the Most Preferred Anti‐Diabetic Drug
`
`“What we were expecting
`was
`that
`by
`the
`time
`omarigliptin was ready to be
`approved
`that
`DPP‐4
`inhibitors would be used as
`first class therapy in the U.S.
`That,
`however,
`has
`not
`happened and Metformin
`continues to be used as first
`line therapy in the U.S.”
`
`Source: EX1073 (Weber Depo. Trans.), 105:15‐20;
`EX1041 (Tanenberg Decl.), ¶¶20‐21; Pet. Reply at 23.
`
`45
`
`
`
`Vildagliptin was Invented Before Saxagliptin
`
`Vildagliptin
`
`Source: EX2013 (Villhauer 2000), 7:15‐27; Pet. Reply at 14.
`
`46
`
`
`
`Vildagliptin Received Regulatory Approval Before Saxagliptin
`
`“Galvus (vildagliptin) , an oral
`DPP‐4 inhibitor, and Eucreas, a
`single‐pill
`combination
`of
`vildagliptin and metformin, are
`indicated for the treatment of
`type 2 diabetes. The products
`were first approved in 2008.
`Galvus is currently approved in
`more than 100 countries[.]”
`
`Source: EX1048 (2012 Novartis Annual Report) at 0039; Pet. Reply at 23.
`
`47
`
`
`
`Sitagliptin was FDA‐Approved Before Saxagliptin
`
`Source: EX1073 (Weber Depo. Trans.), 25:15‐22; EX1028 (Lenhard Depo. Trans.), 397:23‐25.
`
`48
`
`
`
`Sitagliptin is the Highest Selling Gliptin
`
`Source: EX1073 (Weber Depo. Trans.), 27:3‐10.
`
`49
`
`
`
`Saxagliptin Under‐Performs Sitagliptin (Januvia®)
`
`Kombiglyze® XR
`Onglyza®
`
`Saxagliptin products failed to capture a substantial
`share of the U.S. DPP‐4 inhibitor market.
`
`Source: EX1060B (McDuff Decl.), ¶20; EX1062B (McDuff Attachments) at B‐3; EX1035,
`¶36; EX2117 (IMS Health Audit); EX1029 (Meyer Depo. Trans.), 422:4‐13.
`
`50
`
`
`
`Large and Increasing Rebates for Saxagliptin
`
`Onglyza®
`
`2009
`
`2010
`
`2011
`
`2012
`
`2013
`
`2014
`
`2015
`
`Net Sales Adjustments
`as a % of Gross Sales
`
`11%
`
`32%
`
`33%
`
`37%
`
`35%
`
`54%
`
`66%
`
`Kombiglyze® XR
`
`2010
`
`2011
`
`2012
`
`2013
`
`2014
`
`2015
`
`Net Sales Adjustments as
`a % of Gross Sales
`
`26%
`
`33%
`
`37%
`
`38%
`
`53%
`
`57%
`
`Net sales adjustments are a large percentage of gross
`sales and have increased continuously and significantly
`from launch through 2015.
`
`Source: EX1060B (McDuff Decl.), ¶41; EX1062B (McDuff Attachments) at B‐8a; EX1035, ¶63.
`
`51
`
`
`
`Large Marketing Expenditure on Saxagliptin
`
`$571 million in promotional expenditures on Onglyza and
`Kombiglyze XR in the U.S. from 2009 to 2015, which is
`23.0% of total U.S. sales over the same time period.
`
`Source: EX1060B (McDuff Decl.), ¶25; EX1062B (McDuff Attachments) at B‐4.
`
`52
`
`
`
`Saxagliptin Also Underperforms Vildagliptin (Galvus)
`
`Sitagliptin
`
`Vildagliptin
`
`Saxagliptin
`
`Worldwide sales of vildagliptin beat
`worldwide sales of saxagliptin.
`
`Source: EX1060B (McDuff Decl.), ¶18; EX1062B (McDuff Attachments) at B‐1b.
`
`53
`
`
`
`Saxagliptin Also Underperforms Vildagliptin (Galvus)
`
`Vildagliptin (non-US)
`
`Saxagliptin (total sales)
`
`Saxagliptin (non-US)
`
`Even without U.S. sales, worldwide sales of
`vildagliptin beat worldwide sales of saxagliptin.
`
`Source: EX1060B (McDuff Decl.), ¶18; EX1062B (McDuff Attachments) at B‐1c.
`
`54
`
`
`
`They Call Vildagliptin a “Failure”
`
`Source: EX1073 (Weber Depo. Trans.), 23:8‐16, 28:7‐10; EX2057 (Lenhard Decl.), ¶76.
`
`55
`
`
`
`Best‐in‐class Glyptin “a Failure”
`
`Dr. Weber’s own “best in class of the
`gliptins” omarigliptin qualifies as
`an FDA‐approval “failure.”
`
`Source: EX1073 (Weber Depo. Trans.), 99:9‐100:20.
`
`56
`
`
`
`FDA‐Approved DPP‐4 Inhibitors are Interchangeable
`
`Source: EX1028 (Lenhard Depo. Trans.), 402:2‐14; EX1041 (Tanenberg Decl.), ¶¶24‐25; Pet. Reply at 21. 57
`
`
`
`FDA‐Approved DPP‐4 Inhibitors are Interchangeable
`
`Source: EX1028 (Lenhard Depo. Trans.), 398:12‐19, 399:4‐7; EX1041 (Tanenberg Decl.) , ¶¶ 25‐29.
`
`58
`
`
`
`FDA Requires Safety Warning on Saxagliptin Product Label
`
`“Saxagliptin and alogliptin may increase the
`risk of heart failure, particularly in patients
`who already have heart or kidney disease. . . .
`As a result, we are adding new warnings to
`the drug labels about this safety issue.”
`
`Sitagliptin does not have
`a heart failure warning.
`
`Source: EX1032 (FDA Drug Safety Communication);
`EX1041 (Tanenberg Decl.), ¶26; EX1060 (McDuff Decl.), ¶35.
`
`59
`
`
`
`Vildagliptin Also Has a Once‐Daily Combination Form
`
`“When used in dual
`combination with a
`sulphonylurea, the
`recommended dose of
`vildagliptin is 50 mg once
`daily administered in the
`morning.”
`
`Source: EX2080 (Galvus Summary of Product Characteristics) at 3; Pet. Reply at 22.
`
`60
`
`
`
`Additional Slides
`
`Additional Slides
`
`61
`
`
`
`Patent Owner’s Cyanopyrrolidine Arguments
`
`“But there were additional reasons to dismiss
`Compound 25 as a lead. For instance, the cyano group
`introduced the concern of toxic cyanide release in vivo
`(Ex. 2056, ¶162), leading Merck to dismiss
`cyanopyrrolidine compounds[.]”
`
`Source: Patent Owner Resp. at 15, 23‐25.
`
`62
`
`
`
`Cyanopyrrolidines Do Not Release HCN
`
`Source: EX1073 (Weber Cross‐Examination), 52:1‐8.
`
`63
`
`
`
`Cyanopyrrolidines Do Not Release HCN
`
`Source: EX1073 (Weber Depo. Trans.), 52:9‐15.
`
`64
`
`
`
`Ashworth I Table I Explores Optimal N‐Terminal Residues
`
`“To establish an optimal N‐terminal
`residue, we prepared a series of
`amino acid pyrrolidides[.]”
`
`Source: EX1007 (Ashworth I) at 1165.
`
`65
`
`
`
`Bulky Cylcohexyl at P2 Was Most Potent
`
`“In particular, β‐branched α‐amino
`acid derivatives were the most potent
`compounds
`with
`the
`non‐
`proteinogenic
`amino
`acid,
`(S)‐
`cyclohexylglycine providing the most
`active
`pyrrolidide
`(compound
`5
`possessing a Ki value of 64 nM). We
`then applied these findings to a series
`of 2‐cyanopyrrolidides.”
`
`Source: EX1007 (Ashworth I) at 1165.
`
`66
`
`
`
`Ashworth I Table I SAR Applied to Table II
`
`“The series of dipeptide nitriles
`described in Table ll were prepared
`….We were gratified to find that these
`compounds were potent inhibitors of
`DP‐IV. The S.A.R. for the N‐terminal
`residue developed in the pyrrolidide
`series
`correlated well
`for
`the
`dipeptide nitrile series and the most
`potent compounds 24, 25, 26 and 27
`possessed activity comparable to the
`boroprolines, 1 and 2.”
`
`Source: EX1007 (Ashworth I) at 1165.
`
`67
`
`
`
`Ashworth I Finds Excellent Stability Greater Than 48 Hours
`
`“Stability studies revealed excellent
`half‐lives (t1/2)
`in aqueous solution
`(pH 7.4) at room temperature (Table
`II) with several examples having (t1/2)
`greater than 48h.”
`
`Source: EX1007 (Ashworth I) at 1165.
`
`68
`
`
`
`Ashworth Promises Further Investigation at P1
`
`“Further work on optimization of the
`pyrrolidine ring will be reported shortly.”
`
`Source: EX1007 (Ashworth I) at 1165.
`
`69
`
`
`
`Ashworth II Supports Selection of Compound 25
`
`“We recently reported a series of
`aminoacyl‐2‐cyanopyrrolidides6'7
`which possess Ki values of less than 5
`nM versus human DP‐IV and half‐lives
`(t1/2) of greater than 48h in aqueous
`solution (pH 7.4).”
`
`Source: EX2001 (Ashworth II) at 2745.
`
`70
`
`
`
`Ashworth II Seeks Even Greater Potency
`
`“In a quest to improve the potency
`of this class of inhibitors, we
`investigated replacing the
`pyrrolidide ring with other nitrogen
`heterocycles.”
`
`Source: EX2001 (Ashworth II) at 2746.
`
`71
`
`
`
`Ashworth II Supports Selection of Compound 25
`
`“From a range of compounds with
`various heteroatoms
`in 5‐ or 6‐
`membered rings, we were pleased to
`find that
`the 4‐cyanothiazolidide
`analogue 3 was approximately 5‐fold
`more
`active
`than
`the
`2‐
`cyanopyrrolidide inhibitor 57 (Table
`I). However, this increase in activity
`was accompanied by a slight decrease
`in stability.”
`
`Source: EX2001 (Ashworth II) at 2746.
`
`72
`
`
`
`Many Structures Successfully Inhibit DPP‐4
`
`Source: EX2246 (AstraZeneca Demonstrative Exhibit).
`
`73
`
`
`
`In Vivo Toxicity Testing Not Required
`
`Source: EX2174 (Rotella Depo. Trans.), 56:3‐9; Institution Dec. at 11.
`
`74
`
`
`
`Double Bonds Affect Pyrrolidine Conformation Differently
`
`Source: EX2174 (Rotella Depo. Trans.), 124:23‐125:10.
`
`75
`
`
`
`Dr. Rotella’s Prior Art Search
`
`“Q. What I'm trying to get at is how
`you came to know of the existence of
`the materials that you principally rely
`upon in your report.”
`
`“THE WITNESS: I was asked by
`counsel to provide…opinions on
`obviousness related to the
`discovery[,] synthesis[,]
`identification[,] and discovery of
`saxagliptin. And I carried out a
`literature search for documents that
`might be rel[evant]…. In some cases, I
`could ‐‐ I could get those references
`easily; in other cases, I could not.”
`
`Source: EX2174 (Rotella Depo.), 22:15‐26:4.
`
`76
`
`
`
`Dr. Rotella’s Prior Art Search
`
`“[C]an you identify the publications of which
`you had preexisting knowledge prior to the
`commencement of the formulation of your
`opinions?
`A. Ashworth[I]…, Cheng,… Chiou,… Hanessian,
`Korfmacher,… the Villhauer patent….”
`
`“Q. Were there other materials you were
`provided when you began working on this
`program at BMS beyond those two, if you
`recall?
`A. Well, I'm reasonably certain that all of the
`ones I identified were part of the literature
`that we were provided with as a part of
`working on that program…. I apologize. The
`Chiou paper and the Cheng and Prusoff
`paper,
`I was aware of
`from my days in
`graduate school.”
`
`Source: EX2174 (Rotella Depo.), 23:25‐24:22, 25:20‐26:6 .
`
`77
`
`
`
`Dr. Rotella’s Prior Art Search
`
`“Q. No, no, no. At the time that was the subject
`of your report, before the invention of saxagliptin
`was made.
`A. Okay. So to be clear, you're talking about the
`late 1990s? Because my report was written in
`2016, and so I want to make certain that we have
`our date references because ‐‐ the reason I say
`that is because drug discovery has evolved in the
`period of time.
`Q. Well, which period of
`referencing in your report?
`A. I was using the priority dates as provided to me
`by counsel.”
`
`time were you
`
`Source: EX2174 (Rotella Depo.), 58:2‐22.
`
`78
`
`
`
`Dr. Rotella
`
`“Q. And over under the heading
`Conclusion, in the second sentence
`you stated that,
`‘this
`template
`confers substantial
`improvement in
`PDE
`selectivity
`and
`potency
`compared to sildenafil.’ Those were
`your words there, right?
`
`A. Those are – those are my words.
`The substantial
`improvement refers
`primarily, of course, to selectivity.”
`
`Source: EX2221 (Rotella Depo. Transcript), 37:6‐13.
`
`79
`
`
`
`CERTIFICATE OF SERVICE
`
`On 23 January 2017, Mylan served a copy of its demonstrative exhibits on
`
`counsel for Patent Owner AstraZeneca AB at the following electronic service
`
`addresses:
`
`Charles E. Lipsey
`
`charles.lipsey@finnegan.com
`
`Eric E. Grondahl
`
`egrondahl@mccarter.com
`
`John D. Livingstone
`
`john.livingstone@finnegan.com
`
`Anthony A. Hartmann
`
`anthony.hartmann@finnegan.com
`
`M. David Weingarten
`
`david.weingarten@finnegan.com
`
`Nicole A. Conlon
`
`nicole.conlon@finnegan.com
`
`Daniel M. Silver
`
`dsilver@mccarter.com
`
`Kassandra M. Officer
`
`kassandra.officer@finnegan.com
`
`on Wockhardt at the following service electronic addresses:
`
`Frederick R. Ball
`
`FRBall@duanemorris.com
`
`Patrick C. Gallagher
`
`PCGallagher@duanemorris.com
`
`on Teva at the following service electronic addresses:
`
`Iain A. McIntyre
`
`IMcIntyre@carlsoncaspers.com
`
`Gary J. Speier
`
`GSpeier@carlsoncaspers.com
`
`
`
`-1-
`
`
`
`on Aurobindo at the following service electronic addresses:
`
`Sailesh K. Patel
`
`SPatel@schiffhardin.com
`
`Joel Wallace
`
`George Yu
`
`JWallace@schiffhardin.com
`
`GYu@schiffhardin.com
`
`
`and on Sun/Amneal at the following service electronic addresses:
`
`Samuel S. Park
`
`spark@winston.com
`
`Andrew R. Sommer
`
`asommer@winston.com
`
`Respectfully submitted,
`
`
`
`
`By: /Richard Torczon /
`
`Richard Torczon, Backup Counsel
`
`Reg. No. 34,448
`
`
`
`
`
`
`
`Dated: 23 January 2017
`
`
`
`
`
`-2-
`
`