`Editor: H. Timmerman
`
`Volume 12
`
`TRE S IN
`MEDICINAL CHEMISTRV '88
`
`Proceedings of the Xth International Syrnposiurn on Medicinal Chernistry,
`Budapest, 15-19 August 1988
`
`Edited by
`H.VAN DER GOOl
`Department 07' Pharmacochemistry, Vrije Universiteit, Amsterdam, The Netherlands
`'"
`G.DOMANY
`
`Chemical Works of Gedeon Richter Ltd, Budapest Hungary
`
`l.PALLOS
`
`Egis Pharmaceuticals, Budapest, Hungary
`
`H, TIMMERMAN
`
`Department of Pharmacochemistry, Vrije Universíteit, Amsterdam, The Netherlands
`
`ELSEVIER - Amsterdam - Oxford - New York -- Tokyo 1989
`
`NCI Exhibit 2029
`Page 1 of 22
`
`
`
`ELSEVIER SCIENCE PUBLlSHERS B,V,
`Sara Burgerhartstraat 25
`P,O, Box 211, 1000 AE Amsterdam, The Netherlands
`
`Oistributors for the United States and Canada:
`
`ELSEVIER SCIENCE PUBLlSHING CaMPAN Y INC,
`655, Avenue of the Americas
`New York, NY 10010, U,S.!\',
`
`ISBN 0-444-87380-5 (Vol. 12)
`ISBN 0-444-41564-5 (Series)
`
`© Elsevier Sc/ence Publishers B,V" 1989
`
`Al! rights reserved, No part of this publication may be reproduced, stored in a retrieval system or
`transmitted in any form or by any mean s, eleetronic, mechanieal, photoeopying, reeording or
`otherwise, without the prior written permission of the publisher, Elsevier Seience Publishers B, V,I
`Physieal Sciences & Engineering Oivision, P.O. Box 330, 1000 AH Amsterdam, The Netherlands,
`
`Special regulations for readers in the USA - This publieation has been registe red with the Copyright
`Clearance Center Ine, (CCC), Salem, Massachusetts. Information can be obtained from the CCC
`about conditions under which photocopies of parts of this publieation may be made in the USA. Al!
`other copyright questions, ineluding photocopying outside of the USA, should be referred to the
`publisher.
`
`No responsibility is assumed by the Publisher for any injury and/or damage to persons or property
`as a matter of products liability, negligenee or otherwise, or from any use or operation of any meth(cid:173)
`ods, products, instruetions or ideas contained in the material herein. Because of rapid advances in
`the medical sciences, the Publisher recommends that independent verification of diagnoses and
`drug dosages should be made.
`
`Printed in The Netherlands
`
`NCI Exhibit 2029
`Page 2 of 22
`
`
`
`H. van der Coo!, G. Domány, L. Pallo8 and H. Timmerman (Editms)
`Trends in Medicinal Chemistry 'SS
`<D 1989 Elsevier Science Publishers B.V., Amsterdam - Printed in The Ne!hedands
`
`709
`
`EXCITING DEVELOPMENTS IN THE AREA OF HMG-CoA REDUCTASE
`INHIBITORS
`
`F. G. Kathawala, Preclinical Research Department, Sandoz Research Institute, Route 10, East
`Hanover, New Jersey 07936, U.S.A.
`
`SUMMARY
`Investigations by Akira Endo with compactin, a potent inhibitor ofHMG-CoA reductase,
`have to be largely credited for the resurgence of Ihe research on cholesterol biosynthesis and the
`search for novel HMG-CoA reductase inhibitors. HMG-CoA reductase catalyzes the conversian
`ofHMG-CoA to mevalonate, which is an early and rate-limiting step in the biosynthesis of
`cholesterol. Considerations of the structural elements of Ihe substrate (~-Hydroxy-~-Methyl
`Glutaryl-CoA) and compactin involved in interactions at the active site of the enzyme, have
`guided Ihe efforts at Sandoz Research Institute towards the development of a variety of novel
`HMG-CoA reductase inhibitors. Synthesis and Structure Activity Relationships (SAR) of these
`novel inhibitors are discussed with emphasis on the clinical candidate XV 62-320: [R >1< ,S>I<_(E)](cid:173)
`(±)-Sodium-3,5-dihydroxy-7-[3-(4-fluorophenyl)-I-( l-methylethyl)-l H-indol-2-yl]-hept-6-
`enoate, a mevalonic acid analogue more potent than compactin and Lovastatin.
`
`INTRODUCTION
`Coronary heart disease is responsible for approximately 500,000 deaths each year in the
`United States and is associated with direct and indirect cost of more than $60 billion ayear (Ref.
`1). A large body of clinical and epidemiological data has linked elevation of blood cholesterol
`levels as a major cause of coronary heart disease. It has been established that lowering of
`low-density lipoprotein (LDL) cholesterollevels will reduce Ihe risk of coronary heart disease in
`men with elevated blood cholesterollevels.
`The need for the development of effective and safe therapeutic agents for the treatment of
`hyperlipoproteinemia has gained considerable support as a result of two important events: (l)
`the results of the Lipid Research Clinic's Coronary Primary Prevention Trial (LRC-CPPT), a
`multicenter, randomized, double-blind study involving 3,806 asymptomatic middle-aged men in
`the United States with type II hyperlipoproteinemia, which demonstrated that a statistically
`significant reduction of 19% in the Tate of fatal plus non-fatal coronary heart disease was
`associated with a 9% decrease in blood cholesterollevels (Ref. 2), and (2) the recommendation
`to treat individuals with blood cholesterol aboye the 75th percentile, which emerged from the
`consensus panel of the December, 1984 NIH Consensus Development Conference on the
`lowering of blood cholesterol to prevent coronary heart disease (Ref. 3).
`In recent years, to achieve this goal of finding effective and safe therapeutic agents lo lower
`LDL-cholesterol, great interest has focused on potent inhibitors of the enzyme p-HydroxY-Il(cid:173)
`Methyl-Glutaryl-CoA reductase (HMG-CoA reductase, EC 1.1.1.34) which controls a key step
`
`NCI Exhibit 2029
`Page 3 of 22
`
`
`
`710
`
`in the cndogenous synthesis of cholestero1. Several studies, both in animnls and hWllans, h,lve
`been reported with HMG-CoA reduclase inhibitors: compactin (Mevastatin), CS-514
`(Pravastatin), mevinolin (Lovastatin, Mevacor®) and Synvinolin (SimvasLatin) (Ref. 4), which
`ate structurally very c\osely related to one another. In order to assess fulIy the potential of
`HMG-CoA reductase inhibitors as an effective therapeutic intervention for the treatment of
`hyperlipoproteinemia, it is thus desirable to study in human s a variety of these inhibitors derived
`fram different structural prototypes which can be distinguished in their overall biological profile
`fram one anotiler. This conceptual framework forms tbe basis j()!. initiating efforts at the Sandoz
`Research Institule for a variety ofHMG-CoA reductase inhibitors witil chemical structures
`different in several respects from compactin, Pravastatin (a hydroxy analogue of compactin),
`Lovastatin (a methyl analogue of compactin) and Simvastatin (a dimethyl analogue 01'
`compactin) and has led to XV 62-320, the first totally synthetic HMG-CoA reduclase inhihitor
`for studies in human clinical trials (pig. 1).
`
`Mevastatin (Compactin)
`
`Lovastatin (Mevinolin)
`
`H
`
`HO~COONa
`OH
`
`H
`8imvastatin (Synvinolin )
`F
`
`OH OH O
`
`0-
`
`O
`
`H
`
`~
`
`~~ H OH
`
`Fig. 1
`
`XU 62-320
`
`Pravastatin (Eptastatin)
`DESIGN ASPECT
`Investigations by Akira Endo with compactin (Ref. 4) have to be largely credited for the
`resurgence of the research on cholesterol biosynthesis and the renewed interest in HMG-CoA
`reductase inhibitors, a field now almost three decades old. While all intensive studies hitherto
`condllcted have been with c\ose1y relatcd metabolítes, sllch as compactin, mevinolin and C8-5l4
`(Pravastatin), derived fram fungal broths, efforts at the Sandoz Research Institllte towards the
`development of new HMG-CoA reductase inhibitors have been based on synthesis, guided by
`the following assumptions:
`a)
`There are two regions at the active site of the enzyme: olle with high specific
`recognition of a 5-carbon unit (C-l to C-5 as shown below) of the ~-OH-~-Methyl-Qlutaryl
`
`NCI Exhibit 2029
`Page 4 of 22
`
`
`
`porlion and rhe other of CoA moicly present in HMG-CoA.
`4
`2
`CoA S l~ ~<"¡.(OH
`)OH o
`o
`
`IIMG-CnA
`
`Fig. ].
`
`7U
`
`Compactin (R = H, Fig. 3), a known inhibitor ol' the enzyme, may be regarded as a
`b)
`lransition state analoglle, when in the open dihydroxy acid formo
`2
`
`4
`
`¡ OH
`
`o
`
`COl1lpactin ( R= H)
`
`Fig.3
`The S-carbon unit of lhe side chain presenl in compactin (Fig. 3) probably occupies the
`same region as the S-carbon unit in HMG-CoA (Fig. 2); the bicyclic A-B-ring system with its
`substituents in compactin (Fig. 3) possibly sits in the same region or very close lo lhe same
`region the CoA portion of the substrale HMG-CoA oceupies at the active site of lhe enzyme.
`However, il is difficull lo see any similarity in strueture between lhe bicyclic-ring system oC
`compactin and CoA, when one examines lhe Slructure of CoA shown bclow (Fig. 4).
`
`N
`
`HO
`
`o
`O
`J..././\.
`o
`1I
`11
`HS~I(VNIJ '\ 'o/l'o/l'of\~Y N~N
`o
`
`fh NH2
`
`N
`
`y
`
`O
`
`OH OH
`
`Coenzyme A
`
`Fig. ti
`
`rOH
`p ........
`HO"" ¡-o
`OH
`
`In ligh! of the aboye a) and b), one hoped that it might possible to prepare interesting
`synthetic inhibitors of HMG-CoA reductase with a very general structure as shown below in
`Fig. 5, with the 5-carbon unil (C-l to C-S) preferably possessing the absolute configurations 01'
`C-3-0H and C-S-OH as present in compactin.
`
`R
`
`7 6 5 4 3 2
`
`¡ OH
`
`~( R¡
`
`O
`
`Rl =H, eH3
`
`Fig.5
`
`NCI Exhibit 2029
`Page 5 of 22
`
`
`
`712
`
`Choice of R and R¡ in Fig. 5 has depended on:
`a)
`Consideration of the elements of structure of CoA.
`b)
`Considerations of the overall shape and assumptions of the importance of substituents
`on<Ring A-B of compactin (Fig. 3), first with molecular models and later with computer
`modelling.
`c)
`Exploiting the knowledge gained in structure activity relat~onships with our own
`Sandoz Research Institute compounds or being reported in literature by outside investigators.
`Efforts with the aboye considerations in mind have led to the development of a variety of
`novel HMG-CoA reductase inhibitors. Synthesis and Structure Activity Relationships (SAR) of
`these novel inhibitors are discussed below with emphasis on the clinical candidate XV 62-320:
`[R * ,S*-(E)]-(±)-Sodium-3,5-dihydroxy-7 -[3-( 4-flllorophenyl)-1 (1- methylethyl)-lH(cid:173)
`indol-2-yl]-hept-6-enoate (Fig. 1), a mevalonic acid analogue more polent than compactin and
`Lovastatin.
`
`CHEMISTRY
`Guided by the conviction tha! the C-3, C-5 dihydroxy acid fragmenl was the key
`pharmacophore necessary for the inhibition of HMG-CoA reductase, our synthetic approach
`towards the synthesis of compounds of generic structllre (Fig. 5) involved:
`A convergent synthesis coupling chiral synthon 1 or racemic or chiral (3R, 5S) C-3,
`a)
`C-5-dihydroxy ester synthon 2 with a variety of aryl or alkyl fragments 3 (Fig. 6), or
`
`R/'..X
`x = P+(PhhZ(cid:173)
`Y = P=O(OR3)2
`
`1
`Rl = Si(t-Bu)Phz
`
`2
`Fig.6
`
`3
`RZ =R3 = Alkyl, Z- =CI, Br
`
`A linear synthesis of the C-3, C-5 dihydroxy acid derivatives wherein the aldehyde 4
`b)
`is reacted with acetoacetate 5 lo provide a hydroxyketo ester intermediate, which with
`subsequent steps, gives Ihe desired final products of Fig. 5.
`
`R 7 6 5 O
`~
`H
`
`4
`
`Fig.7
`
`5
`
`SYNTHESIS OF SYNTHON 1 AND 2, FIG. 6 (SCHEME 1 AND SCHEME 2)
`Synthon 1 has been synthesized starting from D-glucose via the key líthium alllminllm
`hydride redllctive opening of the epoxide as depicted in Scheme 1 (Ref. 5). The desired axial
`
`NCI Exhibit 2029
`Page 6 of 22
`
`
`
`alcohol could be separated fmm the equatorial isomer by preparation of Ihe silyl derivatives.
`The proteeted axial alcohol on PCC oxidation gave the desired lactol aldehyde.
`Synthesis of chiral Synthon 2 has been aeeomplished starting from S-Malic acid in
`cxcellent yie1ds via an eight-step reaetion as illustrated in Scheme 2 (Ref. 6).
`
`713
`
`LAH
`
`THI<'
`
`HO
`
`OCH3
`
`I RSiCI
`
`OCH3
`
`Irrlidazole
`,
`OCPh3
`OCPh3
`_ OH
`r-..... J:.O
`+ RO ~O
`~O <111 Na/NH3 or
`r---i
`CF3COOH / ~
`RO
`OCH3
`RO
`OCH3
`121P113P~ ~ CHO
`
`CH2cI2
`
`OCH3
`
`lmldazole ~O
`
`~O
`
`OR OCH)
`
`RO
`
`OCHJ
`
`R = Si(t-Bu)Ph2
`
`SCHEME 1
`
`1 • MeOH / AcCI
`
`2. OBMS / NaBH4
`-------Do-
`
`Ph3CO~OCH3
`cm O
`1t -Butylacetate
`
`LOA
`
`o
`H~Ot-Bu
`2 • TBOPSCI
`C)R C)R o
`IMIDAZOLE
`
`R = S i ( t - n u ) Ph 2
`
`<111
`3 • TFA /H20 I CH2C1!
`4 . I'CC I CH2CI2
`
`SCHEME 2
`
`On the other hand, an effieient route was developed for Ihe preparation of racemic Synthon
`2 starting fmm 1,3,5-trihydydroxy benzene through a five-step rcaction sequence shown in
`Scheme 3 (Ref. 7).
`CHOICE OF R AND SYNTHESIS OF INTERMEDIA TES 3, FIG. 6 AND 4, FIG. 7
`Our initial efforts at the synthesis, and the biological results of C-3, C-5-dihydroxy acid
`derivatives (Fig. 5) wherein choice of R was based on elements of substructures of cocnzyme A
`(Fig. 4) or the decalin ring s!mcture of compactin (Fig. 3) were no! promising (Ref. 8). This led
`us to question lhe importance and lhe necessity of the eomplex stereochemistry and the
`substituents present in the decalin ring of compactin and turn our attention towards the
`
`NCI Exhibit 2029
`Page 7 of 22
`
`
`
`714
`
`60H
`
`HO/'
`
`OH
`
`H2IJbNi
`.... - - - - - -P -
`EtOfI
`011
`
`ÓO
`
`:--..
`
`HO
`
`H
`I
`
`OR
`
`OR
`
`0COOCIl1
`
`RO~CHO
`
`l.TFAIM"OH
`~.
`2.PCC/CH2CIZ
`
`RO
`
`TBDPSCI/D~ D
`1~~~CI2
`~
`MCPBA D
`
`NaIlCO]
`CH),CI2
`
`no
`
`O
`
`IMIDAZOl,E
`
`ItO
`
`OH
`
`OR
`
`OR
`
`( '1
`°
`° -<1111-----
`
`(R =TBDPS)
`
`SCHEME3
`
`preparation of C-3,C-5-dihydroxy acid derivatives (Fig. 5) wherein R was a naphthalene ringo
`During these ongoing efforts, we were being encouraged and helped by two importanl
`publications (Rcf. 9) describing rnevalonolactone derivatives of the general structure 6 and 7 as
`inhibitors of HMG-CoA reductase.
`
`O:~"QR,
`R4
`
`y,~ c::~f'
`
`R,¡
`
`6
`
`7
`
`Fig.8
`Further exploration of R in Fig. 5 Icd to the firsl interesting indolyl derivative (Fig. 9)
`comparable lo cornpactin in its inhibitory activity against HMG-CoA reductase.
`F
`
`0-
`Na+
`
`Fig.9
`
`An extensive and rapid analogue program allowed Ihe choice of XU 62-320 (Fig. 1) as a
`candidale for extensive biological testÍng. Currenlly, XU 62-320 is in clinical Phase Il trials.
`With the discovery 01' XU 62-320, the stage was set for a large numher of variations of R in
`Fig. 5. Extensive work al the Sandoz Research InstÍtute has led to rnany novel HMG-CoA
`reductase inhibitors, sorne of which are discussed in this paper and shown in Hg. 10 (Ref. 10).
`Synthesis of the rnany interesting fragrnents 3 (Hg. 6) and 4 (Fig. 7) needed for synthesis of
`final HMG-CoA-R inhibitors are described in Schernes 4-12 below (Ref. 10).
`
`NCI Exhibit 2029
`Page 8 of 22
`
`
`
`7lií
`
`x= CH=CH, (CH2) m
`
`OH OH O
`,
`,
`11 Na+
`Z=~O-
`
`Fig .10
`
`Eithcr one 01' two of the
`aloms a,b,c, = Hetemalom
`
`SYNTHESTS OF INDOLE INTERMEDIATES
`Scheme 4 describes the preparation of 0., ~-unsaturated aldehydes readily obtained fmm a
`variety of 3-phenyl substituted indoles using dimethylamino-acroleín and phosphomus
`oxychloridc while the triphenyl phosphonium salts of indolyl derivatives are prepared vía Ihe
`2-forrnyI and 2-hydmxymethyI indoles using standard procedures (Se heme 5) (Ref. lOa).
`
`Rl~ +
`V NH
`1
`R2
`
`R¡ _
`
`\
`
`/,
`
`~/\;'CHO
`N
`I
`R2
`
`_E_t:_H_-lII>¡" "ON~
`r
`
`CHO
`
`1. Me2N
`POCl3
`
`2. NaOH
`
`SCHEME 4
`
`3 \ /
`
`R¡-- ;: I
`
`\
`N
`
`I
`
`POCl)
`------+
`DMF
`
`QER3 _
`¿p-
`_~~2 _
`
`3\;,
`
`CHO
`
`-ü"'E-
`R¡ ;: I
`\
`N
`I '
`
`NaIlH~4 t ~~ \ ~
`
`I
`
`\
`N
`1
`R2
`
`CH20ll
`
`\
`
`/
`
`R
`¡
`
`-
`:---
`
`I
`
`1. SOCI2
`-
`R
`~---- ¡:---
`2 • P Ph3 I CCI4
`
`+
`H2PPb3
`er
`
`\
`N
`1
`SCHEME 5
`R2
`SYNTHESIS OF INDENE INTERMEDIATES
`A variety of indenyl-a, ~-unsaturated aldehydes and phosphonates have been synthesized
`vía a six-step rcactíon sequence as depícted in Schemes 6 and 7. The synthesis of these
`derivatives involves the preparation of the desired indenes fmm the respective indanones
`
`NCI Exhibit 2029
`Page 9 of 22
`
`
`
`716
`
`folIowed by elther fonnylation at C-2 and subsequent alkylations at C-I or vice versa and then
`proeessing the formyl group through standard reaction sequenees to the desired intennediates
`(Ref.lOb).
`
`t.R2 MgX, HOAc
`
`~
`2. PhN (Me )CHO
`
`POCl3
`
`SCHEME 6
`
`NaH
`
`RtBr
`
`NaBH4
`
`2 SOCl2
`...
`3 P(OMe)3
`
`SCHEME 7
`
`I Ph N(Me)CHO
`
`.. POCl3
`
`R2
`
`~ o
`.. ~
`
`'"
`,,1
`
`11
`P(OMe)2
`
`.R3
`
`RI
`
`"1
`
`RI
`
`SYNTHESIS OF NAPHTHALENE INTERMEDIA TES
`For the preparation of naphthalene derivatives, a novel photochernieal route (Ref. 11) was
`exploited to give the key hydroxy aldehyde whieh on dehydration provides the ene aldehyde.
`Dehydrogenation of the ene aldehyde and ehain extension of the fonnyl group then leads to the
`desired ct, ~-unsaturated aldehydes (Ref. lOe) (Scheme 8).
`
`SYNTHESIS OF IMIDAZOLE INTERMEDIATES
`Highly substituted irnidazole derivatives with the desired funetional group at the desired C(cid:173)
`or he tero- atom are not well described in the literature. Synthesis of the required irnidazole
`intennediates was best accomplished starting from the respective glycine derivatives as shown
`
`NCI Exhibit 2029
`Page 10 of 22
`
`
`
`717
`
`~CHO
`
`88 % by G.c.
`
`H
`
`Florisil
`
`I C6 HS CH3
`
`... Rellux
`
`o
`
`H
`
`l. DDQ , Tol"ene
`
`Z. n-Bu3SnCH=CHOEt
`n -BuLl, THF ,_600
`
`3. p-TsOH ,THF,H20
`16 hrs, r . t_
`SCHEME8
`
`:-..
`
`O
`
`~I
`
`~ I
`
`'"
`
`H
`
`in Scheme 9. The key slep in the synlhetic pathway involves oxidatiol1 01' the methyI group wirh
`potassium persulfale to give the 5-foffilyI imidazo1e derivatives, which through standard
`reaction sequences, give Ihe needed a., p-unsalurated aldehydes or the phosphonates (Ref. lOd).
`
`1. R2COC1
`
`2. AC20 / AcOH, Pyridine, /),.
`
`3. R3NH2' pTsOH, MgS04, Toluene, /),.
`
`CH3 Rl;WR3
`N=i..
`
`R2
`
`4. PC1S ' CHCI3
`
`K2S208/ CuS04' SH20
`CH3Cr\' IH20
`
`RI' RZ' R3 = ALKYL OR SUBSTITUTED ARYL
`
`SCHEME9
`
`SYNTIIESIS OF PYRAZOLE DERIV ATlVES
`A number of pyrazole intermediates have been prepared via procedures dependent on
`whether one needs the 1,5 (Scheme 10), the 1,3 (Scheme 11), or Ihe 3,4 (Seheme 12)
`disubstituted pyrazole intermediares. 2,3-disubstituted pyrazole derivatives are obtained through
`the reaction of the appropriate diketoesters with aryI-hydrazines, needing separation from lhe
`concomitant fOffilation of Ihe eorresponding 1,3 isomer (Scheme lO) (Ref. lOe).
`
`NCI Exhibit 2029
`Page 11 of 22
`
`
`
`718
`
`F
`
`1. LlAIH4l
`
`2.PCC
`
`F
`
`F & ~I
`
`O/
`
`COOEt
`
`O
`
`(
`
`-_ .. _----....
`
`I'hNHNH2
`
`CIl3COOH
`
`F
`
`CHO
`
`1. Ph3P=CHCOOEt
`
`.. 2.LiAIH4
`
`3.Mn02
`
`SCHEME 10
`I ,3-disl1hstituted pyrazoles can be best synthesized from the imide chloride on reaction with the
`acetoacetate derivatives (Seheme 11), while the ring elosure of arylhydrazones give the desired
`
`F :( PC1s ..
`1 ONH
`
`HN
`
`o
`
`F 2 1.
`
`N" el
`
`O~H 2.H+
`/'"
`...-:: SCHEME 11
`
`F
`
`o
`i-Pr· e . CH2COOEt
`
`NaOEt
`
`~
`N,/¡
`N
`
`caOEl
`
`Ó
`
`3,4 diaryl pyrazole intermediates (Scheme 12).
`
`1. ¡.Pr.NHNH2 / EtOH I AcOH (80°)
`~
`
`/1
`
`F 0$,:/1
`:-""lJ, ~
`KOH / diglyrnc J-
`~/I
`
`80°
`
`F
`
`""1
`
`~
`
`\" N-N
`)-
`
`o
`
`SCHEME 12
`
`NCI Exhibit 2029
`Page 12 of 22
`
`
`
`719
`
`S YNTHESIS OF HMG-CoA-R rNHlB lTORS
`AH of the intermediates of the many differenl protolypes described aboye in Schemes 4-12
`cOllld be converted lo the finall-IMG-CoJ\ redllctase inhibitors either llsing lhe linear mute
`involving the "dianion chemistry" or lhe coupllng of the respective phosphonates or
`phosphonium salts with lhe chiral Synthons 1 and 2 (Fig. 6) or with the racemic Synthon 3 (Fig.
`6).
`
`LINEAR ROUTE
`Synlhesis using the linear route is iIlustrated in Scheme 13 for the preparation of the indolyl
`HMO-CoA reductase inhibitors. The key step involves lhe reduction of the hydroxyketoester
`llsing trialkylborane(fHFfMeOH wilh sodium borohydride at _78 0 (Ref. 12) to give lhe mixture
`of dcsired erylhro and threo isomers in lhe ratio of 95-98:5-2%, respectively. In S0l11e cases, the
`boronic esters can be cryslallized which on methanolysis and subsequent hydrolysis wilh sodiUl11
`hydroxide provide the desired sodiUl11 salts. Non-stereoseIective reduction of hydroxyketoester
`with borane t-butylamine complex has been used to prepare a mixture of cis and trans lactones
`separable on flash chromatography (Ref. lOa).
`
`o O
`~OR
`I!>
`NaH I n-BuI.l R ¡
`
`R¡
`
`, CHO
`
`'.B"NH,./
`
`p
`
`OH O
`~COOR
`
`,
`
`2 . NallH4, _780
`
`11. Et3ll,THF
`~'::I ):0
`~ 11 . MeOn
`j' . NaOH, H+
`¿~3 : : I ~OIl NaOH .~3 : : /
`
`R¡
`
`/1 \ ,~COOR
`R
`N I . . . . . N
`
`1
`R2
`
`2. Ca, bodiimide
`3 . ('111 omatography
`
`2
`
`2. NaOH
`
`H
`R,- ~ 1 ~ \.",' O O
`
`OH OH
`---~ R
`COONa
`\ - , ' ,
`/' 1
`1..... N"'--~
`1
`R2
`
`SCHEME 13
`
`CONVEROENT ROUTE
`For illustrative purposes, a convergent route for the preparatiotl of chiral indolyl HMG-CoA
`reductase inhibitors using the sUy! protected Synthon 1 is depicted in Scheme 14. The crucial
`step in this reaction pathway is the oxidation of lactol with RuC12(PPh3h/NMMO (Ref. lOf).
`
`NCI Exhibit 2029
`Page 13 of 22
`
`
`
`no
`
`l. AcOH ITHF IH20
`2. RuCI2 ( "Ph3) 31 NMMO
`
`...
`
`SCHEME 14
`
`R l = Sir t-Bu )1'11 2
`
`In Scheme 15 is shown the use of si1yl protected aldehyde Synthon 2 (derived from malie acid)
`for the synthesis of indenyl HMG-CoA reduelase inhibitors.
`
`LDArrHF
`
`OHC~ OI-lIu
`iT
`~
`§
`OR OR o
`- - - - - .....
`
`/¡. D- BU4NF
`r?j(F
`)V /2.~aOH
`I
`
`ONa
`
`Ol!
`
`F
`
`Ol-Bu
`
`OR OR O
`
`R = Si (t-lIu ) 1'112
`
`SCIIEME 15
`BIOLOGICAL RESULTS AND DISCUSSION
`A 11 initial studies to assess the inhibitory poteney of various eompounds against HMG-CoA
`reductase were eonducted with rat liver mierosomal suspensions, freshly prepared from male
`Sprague-Dawley rats, using an assay for HMG-CoA reduetase activity as deseIibed in
`Aekerman, et al, J. Lipid Res.l1l., 408-413 (1977). The potency of eaeh compound is expressed
`as IC50 (in ~moles, the coneentration which inhibits to the extent of 50% conversion of the
`substrate HMG-CoA lo mevalonate) and for strueture activity relationship compared either to
`compactin = 1 or to XU 62-320 = 1. Below in Tables 1-13 are summarized the most salient
`features of structure activity relationships for a few of the varied structural prototypes as
`HMG-CoA reductase inhibitors being currently studied at the Sandoz Research Institute. In
`Tables 10-13, the Relative Potency column is derived from the lCso values of each compound
`vs. compactin in the in vitro ral microsomal HMG-CoA reductase assay.
`
`NCI Exhibit 2029
`Page 14 of 22
`
`
`
`721
`
`INDOLE DERlV A TIVES
`Table 1 compares the in vitro inhibitory activity against HMG-CoA reduclase of XV 62-320
`with compactin and lovastatin and as their corresponding sodium salts. XU 62-320 is 14ó and
`52- fold more active than compactin and Lovastatin, respectively. As compared to the
`respective sodium salts of compactin and Lovastatin, XU 62-320 is 22 and lO-fold more potenl
`in inhibiting HMG-CoA reductase. 1t is important to note that CUlTent clinical studies are being
`conducted with XU 62-320 which is a dihydroxy acid sodium salt. In contrast, compactin used
`in clinical studies and Lovaslatin (Mevacor®), now marketed, both exist as the lactone forms
`(Fig. 1).
`
`TAULE 1
`
`F
`
`lCso (LlM) Relative Potency*
`
`Compound
`xv 62-320
`146.1
`0.0069
`1.0
`1.011
`Compactin
`2.8
`0.352
`Lovastatin
`6.5
`0.154
`Na Salt Compactin
`0- Na Salt Lovastatin
`14.8
`0.068
`Na+ - -... - - - - - - - - - - - - - - - - - -
`* As compared to Compactin = 1
`Features of the side chain are very important for maximal inhibitory activity as shown in Table
`2. Erythro configuration, as well as Ihe double-bond configuration, are very important
`(anti-isomer 17-fold less active and dramatic 105S of activity for Ihe (Z) diene isomer). The
`dihydro derivative, as well as the ester and the lactone forms, are considerably les s active.
`Maximal inhibitory activity resides in the 3R, 5S antipode.
`
`TABLE2
`
`F
`
`Compound
`xv 62-320
`3R,5S
`3S,5R
`Na Salt, ANTI
`Mcthyl Ester, SYN
`Trans Lactone-(cid:173)
`CIS(Z) Double Bond
`Dih-ydIo (Reduced
`Double Bond)
`* As compared t6 XU 62-320 = 1
`
`lC50 (¡.tM) Relative Potcncy*
`1.0
`0.0069
`2.8
`0.0024
`0.086
`0.08
`0.057
`0.12
`0.13
`0.052
`0.23
`0.029
`0.011
`0.62
`0.06
`0.114
`
`The importance of the features of the side chain described in Table 2 for the indole series holds
`true as well for aJl the prototypes to be described later and hence, during the discussion of SA R
`of these prototypes, these aspects willnot be reemphasized. HMG-CoA, the substrate for the
`HMG-CoA reductase, has at C-3 a methyl group. Tt was important lo determine if an anulogue
`of XU 62-320 canying a methyl group at C-3 would be more poten!. Surprisingly, introduction
`of methyl group at C-3 in either of syn- or anti-configuration \Vas considerably less active
`(Table 3).
`
`Studies of the effects of the substituents in the 3-phenyI ring of Ihe indole moiety are given in
`
`NCI Exhibit 2029
`Page 15 of 22
`
`
`
`722
`
`TAlBLE 3
`
`F
`
`COl11pound
`ICso (~lM) Relative Potency*
`-----------------
`-",,-------
`0.0069
`1.0
`XU 62-320
`R = CH3, 0Y~
`0.049
`O. 14
`O.S 1
`R = CH), ANTI
`0.013
`:JLo- _ _ o
`* As cOl11pared 10 XU 62-320 = I
`
`R OH O
`
`Na+
`
`Table 4. Either electl'on withdrawing or e1ectron donating substituents in the 3-phenyl ring tend
`lo decrease the potency wllich is unaffected by the presence of alkyl groups.
`TABLE4
`
`leso (11M) Relalive Potency"
`
`R
`
`:\ I
`
`O~R
`
`;/
`:-....
`
`41'
`
`~---"--_"
`
`1
`0.049
`1.7
`0.'16
`0.345
`1.38
`0.40
`0.076
`0.006
`
`0.0069
`0.l4
`2-Me
`2-Me,4-F
`0.004
`3-Me 4-F
`0.009
`OH or [ O
`3,S-diMe, 4-F
`0.02
`3,S-diMe
`0.005
`:-.."
`1I
`1" ~'O- H
`0.017
`N
`4-CF
`0.09
`Na+ 4-Scr¡
`1.152
`4-COONa
`>10.0
`- - - - - - - - - - -- - - - - -
`* As cOl11pared lo XU 62-320 = I
`Electron donating or withdrawing substituents (nol shown in Table) Ol' blllky alkyl groups at
`e-5 of the indole moiety led to decrease of potency. Howevcr, alkyl or alkoxy groups al e-4
`anct e-6 tend lo maintain or enhance slightly the polcncy (Table 5).
`
`TABLE 5
`
`F ál
`
`R
`OH OH O
`;/I"~o-
`:-....
`-N
`,
`)__
`Na'
`
`R
`H (62-320)
`4,6-diMe
`4,6-dii-Pr
`5-e6H 11
`60eH2Ph
`
`leso (¡.tM) Relative Potency*
`1.0
`0.62
`1.38
`0.002?
`2.65
`
`0.0069
`0.011
`0.005
`24.0
`0.0026
`
`" As cOl11pared to XU 62-320 = l
`
`Most sensitive lo the activity is the substiluent on Ihe nitrogen of the indole moiety (Table
`6). Optil11al activity 1S provided by the isopropyl group while marked 10ss in potency reslIlts
`with either bulky alkyl 01' phenethyl groups.
`
`Reversing the slIbstituents on N-l and e-3 of the indole moiety to give (Table 7)
`3-isopropyl-N-p-f1uorophenyl analogue of XU 62-320 gives a 4-fold inerease in potency.
`
`Mos! of the substances with a reasonable level of activity against HMG-eoA reductase in in
`vitra microsomal assay were studied in vivo for tlleir effects on inhibitiol\ 01' slerol biosYlllhesis.
`Rcslllts are expressed as EDso (mg!kg), effcclive concentratioll which inhibits to the extcnl 01'
`50% incorporatíon of C l4 acelate into sterols in mts when administered as appropriate doses of
`
`NCI Exhibit 2029
`Page 16 of 22
`
`
`
`TABLE 6
`
`lCso (11M)
`
`Relative Potency*
`
`0.0069
`0.62
`0.096
`50
`49.4
`0.245
`
`1.0
`0.011
`0.071
`0.0001
`0.0001
`0.028
`
`* As compared to XU 62-320 = 1
`
`TABLE 7
`
`R2
`
`OH OH O
`
`~O-,
`
`N
`\
`R¡
`
`Na+
`
`R¡
`
`R2
`i-Pr
`4-FC6H4, syn
`(62-320)
`4-FC6H4
`
`i-PI, syn
`
`ICsoO.l.M)
`
`0.0069
`
`0.0016
`
`i-Pr
`
`4-FC6B 4, anti
`
`0.12
`
`ReJative Potency
`as Compared to
`XU 62-320 = 1
`
`1.0
`
`4.3
`
`0.057
`
`drug slIbstances as compared to controls receiving vehicJe alone. Table 8 shows that in vivo
`XU 62-320 is about 40 and 4.5-fold more potent than compactin and Lovastatin, respectively, in
`inhibiting endogenous cholesterol synthesis in rats. For most substances, although not for al!,
`TABLE8
`
`F
`
`Compolll1d
`
`ED50(mg/kg)
`
`Relative Potency*
`
`XU 62-320
`Compactin
`Lovastatin
`(Monacolin)
`
`0-
`
`0.093
`3.5
`0.414
`
`37.6
`1.0
`8.4
`
`* As compared to Compactin = 1
`
`the relative potency dctnmined in in vitro microsomal assay against HMG-CoA reductase
`parallels the in vivo aClivily in rats for the inhibition of 14C-acetate into stcrols. As an example,
`comparison of Tables 2 and 9 reveals the relative potency of several analogues ol' XU 62-320
`when compared in in vitro and in in vivo. Thus, as compared to XU 62-320, the anti isomer is ~
`17- (Table 2) and ~ 15- (Table 9) fold less active than XU 62-320 in in vitro and in in vivo
`assays, respectively. Similarly, close parallelism prevails for the ester (less active ~ 7.5-fold,
`vitro vs. 4.3-fold, vivo), tfans lactone (less active 4.2-fold, vitro vs. 3.5, vivo) and the dihydro
`derivative (less active 16.S-fold, vitro vs. 13-fold vivo).
`
`INDENE DERIV ATlVES
`The strueture activity relationships for the indene derivatives can be best summarized as
`folJows: Maximal activity is obtained with aspiro cyclopentyl group at C-l, again emphasizing
`the importance of the bulky group in the vicinity of the dihydroxy acid side chain. At C-3 the
`
`NCI Exhibit 2029
`Page 17 of 22
`
`
`
`724
`
`TABLE 9
`
`F
`
`Compound
`
`EDso(rng/kg) Relative Potency*
`
`/
`
`1.0
`1.66
`
`~
`
`:-..
`
`0,093
`XU 62·320
`3R, SS
`0.0:i6
`>0.5
`3S, SR
`0.067
`1.37
`Na Salt, Anti
`OH OH
`" ~
`0.23
`0.40
`Methyl es ter, Syn
`. • V
`~\ ~ Trans Lactone--
`0.28
`0.33
`0.075
`0-
`Dihydro (Reduced 1.23
`l~ N
`Na+ ___ D_o_u_b_l_e_B_o_n_d) ____ ~ ________________ __
`) -
`* As compared to XU 62-320 = 1
`XU 62-320
`best substituent is 4-F-phenyl, while the optimal substituent for the benzenoid portion of the
`indene moiety is hydrogen (see Table 10).
`
`TABLE 10
`
`....
`
`F
`
`Na+
`0-
`
`OH OH O
`
`R¡
`
`(CH2)4
`(Racemic)
`(CH2)4
`(3R, SS)
`(CH2h
`(CH2)s
`CH2CH3
`CH}
`H,iPr
`
`Relative Potency
`as Compared to
`Compactin = 1
`
`202
`
`337
`
`38
`1.5*
`< .2
`2
`8
`
`Na+
`O'
`
`(
`OH OH O
`
`88*
`Phenyl
`3,5-Dimethylphenyl 146
`< 0.5
`iPr
`CycJohexyl
`16.5
`
`F
`
`Na+
`0-
`I ----...:"
`... ",~
`OH OH O
`
`R,
`
`4-Me
`6-Me
`7-Me
`6-0Me
`4,6-(OMeh
`
`114
`181
`24
`130
`60
`
`* As its Ethyl Ester
`
`NAPHTHALENE DERlVATIVES
`
`The most interesting part of the structurc activity relationships for lhis group of compounds
`is the difference observed in the potency of 1-4'-F-phenyl-3-isopropyl derivative vs.
`l-isopropyl-34' -F-phenyl compound (22 times more poten! vs. 337 as compared 10 compactin)
`(see Table 11).
`
`NCI Exhibit 2029
`Page 18 of 22
`
`
`
`TABLE 11
`
`R¡
`
`OH OH O
`
`~O-
`~n NN~+
`a
`R2
`
`72.5
`
`Relative Potency
`as Compared lo
`Compactin '" I
`
`0.10
`8
`19
`22
`56
`2
`337
`144
`
`R¡
`
`4-F-Ph
`4-F-Ph
`4-F-Ph
`4-F-Ph
`3,S-diMe-Ph
`Ph
`i-Pr
`i-Pr
`
`H
`CH3
`Et
`i-Pr
`CH3
`CH3
`4-F-Ph
`Ph
`
`PYRAZOLE DERIVA TIVES
`Table 12 ilIustrates the structure activity relationships for a few of the many pyrazole
`derivatives prepared. Here, too, the optimal substituents are the 4-F-phenyI and isopropyl group
`adjacent to the dihydroxy acid side chain. The dihydro or the S-keto derivative are substantially
`less potent. 1,3-diaryl subslituted pyrazole derivatives show decreased inhibitory aClivity (nol
`shown in the lable) in contrast to Ihe 1,5 and 3,4-diaryl substiluted compounds, which tend lo
`have comparable potency.
`
`TABLE 12
`
`Relative Potency
`as Compared lo
`Compaclin '" 1
`
`R
`
`0-
`
`4-F
`4-F (6,7 Dihydro)
`4-F (5 Keto)
`H
`3,5 Dimethyl
`
`4-F
`
`0-
`
`60
`5.9
`3.5
`5.6
`4.1
`
`30
`
`IMIDAZOLE DERIV ATIVES
`To emphasize Ihe mosl salient features of the structure activity relationships for the
`imidazole derivatives, only a few of Ihe derivatives prepared are tabulated in Tab1e J 3. Optimal
`activity is obtained wiLh l,2-diaryl derivatives with the 4-F substituenl preferred in the phenyl
`rlng on nitrogen and H atom being the preferred subslituenl for tbe phenyl ring at C-2. Alkyl
`substituents at C-2 tend to lead lo considerable loss·of activity. The 1,3-diaryl substituted
`imidazole derivalives suffer a drama tic 10ss of activity when compared to the very potent
`1,2-diaryl compounds.
`
`NCI Exhibit 2029
`Page 19 of 22
`
`
`
`726
`
`TABLE 13
`
`F
`
`~
`
`dN\~'R3
`I " --
`
`Relative Potency
`as Compared lo
`Compaclin = J
`
`337
`
`532
`
`84
`20
`7
`10
`
`4-F
`(Racemic)
`4-F
`(3R,5S)
`p-Cl
`p-Br
`3,5-Di-Me
`3,5-Di-Cl
`
`i-Pr
`t-Butyl
`eyc10hexyl
`2-Thienyl
`1,4-Biphenylyl
`p-Dimethylamino-phen y 1
`p-Nitro-phenyl
`
`4.4
`4.4
`4.8
`202
`35
`56
`72
`
`0-
`Na+
`
`i-Pr
`4-F-Phenyl
`
`1.1
`< 0.1
`
`CONCLVSION
`This paper has described exlensive efforts at Sandoz Researeh Institute towards the
`synthesis and structure aetivity relationships of a variety of HMG-CoA reduetase inhibitors.
`These many prototypes, in addition lo a host of others hitherto not yet described, continue to be
`studied extensively in our laboratories for their effects not only on HMG-CoA reduetase but also
`on the lipoprotein levels in severa! species. While XV 62-320 continues to be studied in clinical
`mals, further invesligations with several Sandoz HMG-CoA reduclase inhibitors hope to
`provide an assessmenl of their true potential and safety for the treatment of elevated Low
`Density Lipoprotein (LDL)-cholesterol1evels.
`
`ACKNOWLEDGEMENTS
`The extensive work described here is truly an outcome of a cohesive team effor! of a very
`large number of dedicated and creative individuals. Most important original contributors