Alnylam Exh. (cid:20)(cid:19)(cid:24)(cid:19)
`
`

`

`Taylor & Francis Group
`
`CRC Press
`
`

`

`CRC Press
`Taylor & Francis Group
`6000 Broken Sound Parkway NW, Suite 300
`Boca Raton, FL 33487-2742
`
`© 2008 by Taylor & Francis Group, LLC
`CRC Press is an imprint of Taylor & Francis Group, an Informa business
`
`No claim to original U.S. Government works
`Printed in the United States of Americaon acid-free paper
`10987654321
`
`International Standard Book Number-10: 0-8493-8796-5 (Hardcover)
`International Standard Book Number-13: 978-0-8493-8796-8 (Hardcover)
`
`This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted
`with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to
`publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of
`all materials or for the consequencesoftheir use.
`
`Nopart of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or
`other means, now knownor hereafter invented, including photocopying, microfilming, and recording,or in any informa-
`tion storage orretrieval system, without written permission from the publishers.
`
`For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://
`www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA 01923,
`978-750-8400. CCCis a not-for-profit organization that provides licenses andregistration for a variety of users. For orga-
`nizations that have been granted a photocopylicense by the CCC,a separate system of payment has been arranged,
`
`Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for
`identification and explanation without intent to infringe,
`
`Library of Congress Cataloging-in-Publication Data
`
`Antisense drug technology: principles, strategies, and applications / editor Stanley T. Crooke. -- 2nd
`ed.
`
`p.jem,
`Includes bibliographicalreferences and index.
`ISBN-13: 978-0-8493-8796-8 (alk. paper)
`ISBN-10: 0-8493-8796-5 (alk. paper)
`1, Crooke, Stanley T.
`1. Antisense nucleic acids--Therapeutic use.
`[DNLM:1. Oligonucleotides, Antisense--therapeutic use. QU 57 A6324 2006]I, Title.
`
`RM666,A564A567 2006
`615’.31--dc22
`
`Visit the Taylor & Francis Web site at
`hitp://www.taylorandfrancis.com
`
`and the CRC Press Website at
`http://www.crepress.com
`
`2006101712
`
`

`

`Preface gagaudvevevesuseaedaanabanauctFenabvedss<d0snastsasnsatersespecyfvecasahassddeasncegerenssidusedeassesksssnsasasseasedseresdesepesqedye ses lk
`
`Contents
`
`The Editor............. we Gsa aU SLs Va nU My iy AC Ra gy PRU Gan a Se cass asT Lie pccuadepiii deeds euNNNTTR SEIN TEDCt xiii
`
`GCONtPADUlONSiscc esi ssapnnse sions ge nace eesied acca sai pandaaspgns nays yesdasonnacngyotie a dewann vevnugy0das) pov indersd taghcas’h ge bead sapaiein vereXV
`
`Part I
`TriPROMUCELON Sescessaasacesna secpciatedvce teas earmet acs ceneantoaenas saans sa sahncka vy Faicgsva sug GuASR Ses E naNaam aRNaR GRRL ep tpmaich l
`
`Chapter 1
`Mechanisms of Antisense Drug Action, an Introduction...
`
`Stanley T. Crooke, Timothy Vickers, Walt Lima, and HongjiangWu
`
`as a
`
`Chapter 2
`"TReRINaSe Hl MSCHATISOG 5: ccapstorcisvaseurseentvavenavas Enea aaneivertaedection athMinis dcmnueion47
`Walt Lima, Hongjiang Wu, and Stanley T. Crooke
`
`Chapter 3
`Sis RNA Silencing Pathways,csvavarsssespronvercvoronvcessiagndseneentteysannbapnsash vanacnuassbacedearevapemi temsadeleaaty>
`Alla Sigova and Phillip D. Zamore
`
`Chapter 4
`Splice Switching Oligonucleotides as Potential Therapeutics ........0.ccccccccccceceesecceeesesseseeeseseseseaceeees9
`Peter Sazani, Maria A. Graziewicz, and Ryszard Kole
`
`PartII
`The Basics of Oligonucleotide-Based Therapeutics -.............eccesecesseeereseeeeeneneneesseaseusscusseneanenvanene 115
`
`Chapter 5
`Basic Principles of Antisense Dine DisGovery icccawunuchonananannakntincimnancanncmuwned Lt
`Susan M.Freier and Andrew T. Watt
`
`Chapter 6
`The Medicinal Chemistry of Oligonucleotides soi icceypntinecccesstseriastes pssst nthaedsviourinanensyess 143
`Eric E. Swayze and Balkrishen Bhat
`
`Chapter 7
`Basic Principles of the Pharmacokinetics of Antisense Oligonucleotide Drugs w......ccccccceceonseeeee 183
`Arthur A. Levin, Rosie Z. Yu, and Richard S. Geary
`
`Chapter 8
`Routes and Formulations for Delivery of Antisense Oligonucleotides ..........0....cccceccececseeseneseeees 217
`Gregory E. Hardee, Lloyd G. Tillman, and Richard S. Geary
`
`Chapter 9
`Liposomal Formulations for Nucleic Acid Delivery......2.c0ccccccccccccseisseeseeseeseeeetseseresnesesientees 237
`Jan MacLachlan
`
`

`

`vi
`
`CONTENTS
`
`Part WIA
`Hybridization-Based Drugs: Basic Properties 2’-O-Methoxyethyl Oligonucleotides................
`
`27 |
`
`Chapter 10
`Pharmacological Properties of 2'-O-Methoxyethyl-Modified Oligonucleotides.........2.-.s-1+0
`C. Frank Bennett
`
`273
`
`Chapter 11
`
`Pharmacokinetic/Pharmacodynamic Properties of Phosphorothioate 2'-O-aeeeeivone
`Modified Antisense Oligonucleotides in Animals and Man..
`:
`este timlanas
`Richard S. Geary, Rosie Z. Yu, Andrew Siwkowski, and Arthur x: Lain.
`
`ives OS
`
`Chapter 12
`Toxicologic Properties of 2'-O-Methoxyethyl Chimeric Antisense Inhibitors
`in Animals and Man...
`noi
`Scott P. Henry, Tae-WonKim,Rinabedty KramepStickland,.
`Thomas A. Zanardi, Robert A. Fey, and Arthur A. Levin
`
`ramen
`
`Chapter 13
`An Overview of the Clinical Safety Experience of First- and Second-Generation
`ADUSSTISS COPONUIGOEDes ag gate eas bce ts cpenapceapcpcesncy opadonisientpas wnaticn otodch ce cbne baa ahah dh chaadeaprebenciebeneves
`T. Jesse Kwoh
`
`365
`
`Chapter 14
`Manufacturing and Analytical Processes for 2'-O-aSteeee
`CHROORUCIEOIOSS oj assatieniaip einer pert needeueundeesFama{auilvies ty coos nairoguiboeauti-eaatearteanvvngadnevosumnmmaanecaixe
`Daniel C. Capaldi and Anthony N. Scozzari
`
`401
`
`Part II B
`Hybridization-Based Drugs: Basic Properties Duplex RNA Drugs...0.......2..2:.c:cccceeeeeeeseeneeeenees
`
`A435
`
`Chapter 15
`Utilizing Chemistry to Harness RNA Interference Pathways for Therapeutics:
`Chemically Modified siRNAs and Antagomirs...
`sanacacssaassessesssacscascasas apaarqaczesemacsonpenes
`Muthiah Manoharan and Kallanthottathil G. Ralcey”
`
`+437
`
`Chapter 16
`Discovery and Development of RNAi Therapeutics ..............ccc:ccseseeeereseneseseensesneeaesesecennenneeeeeenes
`Antonin R. de Fougerolles and John M. Maraganore
`
`465
`
`Part IV
`Coiner CamereCSRen OE LUTig ocaecacksataeal alice onl oat shy sik seeksoabehedpipn eds dake taapadiAicinbndisie suspesdeorchens
`
`Chapter 17
`Optimization of Second-Generation Antisense Drugs: Going Beyond Generation 2.0...
`Brett P. Monia, Rosie Z. Yu, Walt Lima, and Andrew Siwkowski
`
`Chapter 18
`Moadnuilatino Gene Finetion with Pentide Nucleic Acide (PNA\
`
`en AST
`
`S07
`
`

`

`CONTENTS
`
`Chapter 19
`Locked Nucleic Acid...
`Troels Koch and Henriktant
`
`Chapter 20
`Morpholittos...0..-s-++s-stessassisossvenrsssecterntorebenntvsrtos
`Patrick L. Iversen
`
`vii
`
`ere 9
`
`565
`
`Part V
`PicrApSUitle APpCalaOPS sy cpcceam avons a ejarmrn pagent dh dak openan pape Biaukeukt pepipape ved ahncetecaaleoannheh
`
`583
`
`Chapter 21
`Potential Therapeutic Applications of Antisense Oligonucleotides in Ophthalmology...............
`Lisa R. Grillone and Scott P. Henry
`
`585
`
`Chapter 22
`Cardiovascular Therapeutic Applications .........20.cccccccecete eet cceeseeasecoeeteeeuee sacicacatebsseniamsiiaasee
`Rosanne Crooke, Brenda Baker, and Mark Wedel
`
`Chapter 23
`Developing Antisense Drugs for Metabolic Diseases: A Novel Therapeutic Approach...........
`Sanjay Bhanot
`
`641
`
`Chapter 24
`PAUEsUeVEELAUTYT SEGI esvsrvcian oad Since chp CHASE, cap fp natin domhleeearopsdagtinds andmuws tavaizeb phos IR
`665
`Susan A. Gregory and James G. Karras
`
`Chapter 25
`Antisense Oligonucleotides for the Treatment Of Cancer.....,..-...sre::sseseresssseornennrarerssessensugeeeneess
`Boris A. Hadaschik and Martin E. Gleave
`
`Chapter 26
`Targeting Neurological Disorders with Antisense Oligonucleotides ................cscs0c0c0sesecee
`Richard A. Smith and Timothy M. Miller
`
`Chapter 27
`Mechanisms and Therapeutic Applications of Immune Modulatory Oligodeoxynucleotide
`and Oligoribonucleotide Ligands for Toll-Like Receptors ..............-cccccssssseesessrsetessseesensesennerenrs
`Jorg Vollmer and Arthur M.Krieg
`
`Chapter 28
`Aptamer Opportunities aiid Chalenges a: ccasvsasasasspananesapsssteronitnandesvcpaeaisis tadadsbeag ies seta ikiaanasantooosas
`Charles Wilson
`
`699
`
`721
`
`747
`
`773
`
`EAEi a pace am eam Sac CARRE SIN SN TSAATIND Uae oomb LAS SUNTAN GAAP RS nT Soa ATE
`sOl
`
`

`

`Discovery and Development of RNAi Therapeutics
`
`CHAPTER 16
`
`Antonin R. de Fougerolles and John M. Maraganore
`
`CONTENTS
`
`Introduction...
`
`16.1
`
`16.2
`
`16.3.
`
`saniiteSnigile capaaachiloig Soaastocataneine bins oyeepaiteabeaneeiauesiiaes
`
`jiiseptudEnnsaiahaialSanvreiaaaauntaldpaipaiatiopinadSiaaibe
`
`466
`466
`BiianctOl
`vevever 468
`
`In Vitro Selectionof LeadCandidates...
`16.2.1 Potency...
`16.2.2
`Specificity...
`16.2.3
`Stability ...
`és
`sere 68
`16.2.4 TherapeuticConsiderations.
`adostnvendOD
`patel
`In Vivo Delivery ...,
`Shavhaweencsuslgst euatps
`16.3.1 Naked siRNA..
`SetteslaaiructokideastkstboSewaneeierhesedabhatvenemfuse
`sermnargesitLL
`471
`
`:
`
`|
`
`16.3.2. Conjugation...
`
`16.3.1.2 Respiratory...
`
`veveene4] 2
`
`16.3.1.3. Nervous System.
`bagAaNAPohabilcavaeeataadeRTadRaooNasachnatAD
`Sic
`16.3.2.1 Cholesterol ..
`°16.3.2.2 OtherNaturalLigands.
`
`vse 474
`Gaesasanscascviaavha leesTans eae cade iesadetlnda ala vente are raa
`wiT4
`474
`wun
`475
`rauROR Ua hsHh ves ea veh etn TAO FE INTER Qs conan nbaRE NSLS
`16.3.2.3. Aptamers...
`16.3.2.4 Small Molecules.cissssisssssssssiissinsssanssnisivbveeninsoivntusinnsnn
`475
`475
`“Lalisrcies arid Losseo sm sh ycrseonneaetneseacpect ys vestofa leeds obey eaves veeese nana
`P6553,
`16.34 Peptides atid Polymers cars ynecncsnweanerity adeio ea iemavesnneriaerncceeiieite Ganisiens
`478
`DSSS!
`SPR MRRARUGSe aeacrr oeteSpcaieecey leeds nye wees rearsk face cpee eae BTED ir eeer ee
`vvee478
`16.4 ClinicalTrials...
`scutes clan Ueauspsenastana tiotiay Sing ed ane ATP Tea a oD aaa eNRE OsES
`weve 478
`480
`
`477
`
`16.4.1.2 VEGF RSAucdnecceRRRckminuacmeces®E
`480
`480
`GSE:vERRANAGONY 5sige Pasaypcacasnctectienenchant isseaapcnanaahnntyegiiartgnsdewtacphagatgasaiasnnly
`480
`RAEN|AIRS Wie par eesiesangscinineeecierln Giaeg lash hanrineLeeann
`480
`LG:5)
`“SUBIMALY: Mist scaadivistassagssccerassncinaecensansan ibaa saqaninababihscatentestiadsvanceropsthuveds laidudvaisesTuagaanngcimuanel
`4 81
`RETELENCCS sccsexeareactoseeint
`
`465
`
`

`

`466
`
`ANTISENSE DRUG TECHNOLOGY, SECOND EDITION
`
`16.1
`
`INTRODUCTION
`
`In less than a decade since its discovery, RNA interference (RNAi) as a novel mechanism to
`selectively silence messenger RNA (mRNA)expression has revolutionized the biological sciences
`in the postgenomic era, With RNAi, the target mRNAis enzymatically cleaved, leading to decreased
`levels of the corresponding protein. The specificity of this mRNA silencing is controlled very
`precisely at the nucleotide level. Given the identification and sequencing of the entire human
`genome, RNAiis a fundamental cellular mechanism that can also be harnessed to rapidly develop
`novel drugs against any disease target. The reduction in expression of pathological proteins through
`RNAiis applicable to all classes of molecular targets, including those that have beentraditionally
`difficult
`to target with either small molecules or proteins, including monoclonal antibodies.
`Numerous proof-of-concept studies in animal models of human disease have demonstrated the
`broad potential applicability of RNAi-based therapeutics. Further, RNAi therapeutics are now under
`clinical investigation for age-related macular degeneration (AMD) andrespiratory syncytial virus
`(RSV) infection, with numerous other drug candidates poised to advanceinto clinical development
`in the years to come.
`In this review, we will outline and discuss the various considerations that go into developing
`RNAi-based therapeutics starting from in vitro lead design and identification, to in vivo preclinical
`drug delivery and testing, and lastly, to a review of clinical experiences to date with RNAi thera-
`peutics. While both nonviraldelivery of small interfering RNAS (siRNA)andviral delivery of short
`hairpin RNA (shRNA)are being advanced as potential therapeutic approaches based on RNAi,this
`review will focus solely on development of synthetic siRNA as drugs. Synthetic siRNAs can har-
`nessthe cellular RNAi pathwayin a consistent and predictable manner with regard to the extent and
`duration of action, thus making them particularly attractive as drugs. As a consequence, siRNAsare
`the class of RNAi therapeutics that is most advanced in preclinical andclinicalstudies.
`
`16.2
`
`IN VITRO SELECTION OF LEAD CANDIDATES
`
`This section highlights the various steps required to identify potent lead siRNA candidates
`starting from bioinformatics design through to in vitro characterization. The overall scheme for
`turning siRNAinto drugs is summarized in Figure 16.1. The three most importantattributes to take
`into account when designing and selecting siRNA are potency, specificity, and nuclease stability.
`
`5SECSUse 3_Sense '
`
`TRRRARRRARARARRRARRIRP 8
`iri sone
`{SIGE SIE
`Anti-Sense
`F
`1
`»
`In silico design
`»
`In vitro assays
`
`’
`
`Sugar modification
`
`» Antibodies
`
`eae
`
`+ Stabilize siRNA
`» Chemistry
`
`* Select Delivery
`» Naked
`» Conjugation
`» Liposomes
`» Peptides/Polymers
`
`Figure 16,1. Turning siRNAinto drugs. Outline of steps involved in development of an RNAi therapeutic. This
`three-step process begins with jn silico design and in vitro screening of target siRNA,followed by
`incorporation of stabilizing chemical modifications on lead siRNA as required, and ending with
`selection and jn vivo evaluation of delivery technologies appropriate for the target cell type/organ
`and the diseasesetting.
`
`

`

`DISCOVERY AND DEVELOPMENT OF RNAi THERAPEUTICS
`
`467.
`
`With regard to specificity of siRNA,the twoissues of“off-targeting” due to the silencing of genes
`sharing partial homology to the siRNA and “immunestimulation” due to recognition of certain
`siRNAsby the innate immune system have been of special concern. With an increased under-
`standing of the molecular and structural mechanism of RNAi, all issues around lead siRNA
`selection are better understood and also are now generally resolvable through the use of bioin-
`formatics, chemical modifications, and empirical testing. Thus, it is now possible to very rapidly,
`in the span of several months, identify potent, specific, and stable in vitro active lead siRNA
`candidates toany target of interest.
`
`16.2.1 Potency
`
`Work by Tuschl and colleagues [1] represented the first published study to demonstrate that
`RNAi could be mediated in mammalian cells through the introduction of small fragments of double-
`stranded RNA (dsRNA), termed smallinterfering RNA, and that siRNAs hada specific architecture
`comprised of 21 nucleotides in a'staggered 19-nucleotide duplex with a 2-nucleotide 3’ overhang on
`each strand (Figures 16.1 and 16.2). Further elaboration and dissection of the RNAi pathway,
`including: insights from X-ray crystallographic structures, have revealed that long dsRNAsare
`naturally processed into siRNAs via a cytoplasmic RNasellI-like enzymecalled Dicer. A multienzyme
`complex known as the RNA-induced silencing complex (RISC) then unwinds the siRNA duplex.
`The siRNA-—RISC complex functions enzymatically to recognize and cleave mRNAstrands com-
`plementary to the“antisense” or “guide” strand of the siRNA. The target mRNAis then cleaved
`between nucleotides 10 and 11 (relative to the 5’ end of the siRNA guide strand). Loading of RISC
`with respect to the sense‘and antisense siRNAstrands is not symmetrical. Theefficiency with which
`the antisense or guide strand is incorporated into the RISC machinery (versus incorporation of the
`sense strand) is the most important determinant of siRNA potency. Throughanalysisof strand-specific
`
`‘Synthetic
`
`siRNAs WA
`
`
`
`
`
`Cleavage
`
`WAV/\\ siRNAs
`
`
`Dicer
`
`
`
`Therapeutic
`genesilencing
`
`Strand separation
`
`RISC
`
`Natural
`process of
`RNAi
`
`| Complementary pairing
`
`APPLE (An
`mRNA
`
`Cleaved mRNA
`mRNA
`degradation —_— 4\
`
`Cleavage
`
`a”
`
`nN
`
`.™ (An
`
`Figure 16.2 Harnessing the natural RNAi process with synthetic siRNA. Long double-stranded RNA (dsRNA)
`is cleaved into short stretches of dsRNAcalled siRNA. The siRNAinteract with the RISC to selec-
`tively silence target mRNA. siRNA against any mRNAtarget can be chemically synthesized and
`introducedinto cells, resulting in specific therapeutic gene silencing.
`
`

`

`468
`
`ANTISENSE DRUG TECHNOLOGY, SECOND EDITION
`
`it was found that RISC
`reporter constructs [2] and large sets of siRNA of varying potency [3,4],
`preferentially associates with the siRNA duplex strand whose 5’ end is boundlesstightly with the
`other strand. A detailed description of RNAi-mediated silencing as it relates to siRNA and other
`small RNAs by Sigova and Zamore can be found in Chapter 3 of the book.
`
`16.2.2 Specificity
`
`RNAi-mediated silencing of gene expression has been shownto be exquisitely specific as evi-
`denced by silencing fusion mRNA without affecting an unfused allele [5,6] and by studies show-
`ing ability to silence point-mutated genes over wild-type sequence [7]. Nevertheless, along with
`on-target mRNAsilencing, siRNA might have the potential to recognize nontarget mRNA,other-
`wise known as “off-target’ silencing. On the basis of in vitro transcriptional profiling studies,
`siRNA duplexes have been reported to silence multiple genes in addition to the intended target
`gene under certain conditions. Not surprisingly, many of these observed off-target genes contain
`regions that are complementary to one of the two strands in the siRNA duplex [8-10]. More
`detailed bioinformatic analysis revealed that complementarity between the 5’ end of the guide
`strand and the mRNA wasthe key to off-target silencing, with the critical nucleotides being in
`positions 2-8 (from the 5’ end of the guide strand) [11,12]. Accordingly, careful bioinformatics
`design of siRNA can reduce potential off-target effects. Further, published work has shownthat.the
`incorporation of 2'’-O-Me ribose modifications into nucleotides can suppress most off-target
`effects while maintaining target mRNAsilencing [13,14]. In fact, incorporation of a single 2'-O-Me
`modification at nucleotide 2 was sufficient to suppress most off-target silencing of partially
`complementary mRNAtranscripts by all siRNAstested. Thus, in summary, bioinformatics design
`and position-specific, sequence-independent chemical modifications can be incorporated into
`siRNA that reduce off-target effects while maintaining target silencing.
`A second mechanism whereby siRNA can induce potentially unwantedeffects is through stim-
`ulation of the innate immune system in certain specialized immunecell types. It has been demon-
`strated that siRNA duplexes contain distinct sequence motifs that can engage Toll-like receptors
`(TLRs) in plasmacytoid dendritic cells and lead to increased interferon-alpha production [15]. In
`much the same way'that. certain CpG motifs in antisense oligonucleotides are responsible for TLR-
`9-mediated immunostimulation, the interferon induction seen with discrete siRNA nucleotide
`motifs was found to occur largely via TLR-7. Much additional work remains to be donein identi-
`fying the full spectrum of immunostimulatory motifs and whether other receptors might also be
`involved (reviewed in [16]). Several approaches exist to circumvent the immunostimulatory prop-
`erties of certain siRNA duplexes. First, in vitro assays exist to rank-order duplexes for their abil-
`ity to induce interferon when transtected into plasmacytoid dendritic cells [15]. Second, several
`groups have shownthat introduction of chemical modifications, such as 2’-O-Me modifications,
`are capable of abolishing immunostimulatory activity [15,17,18]. Third, siRNA delivery strategies
`can be employed that would avoid the cell types responsible for immunestimulation.
`
`16.2.3 Stability
`
`the chemical modification of siRNA
`Not surprisingly, numerous studies have shown that
`duplexes, including chemistries already in use with antisense oligonucleotide and aptamer thera-
`peutics, can protect against nuclease degradation with no effect or intermediate effects on activity
`[19-21]. For instance, introduction of a phosphorothioate (P=S) backbonelinkageat the 3’ endis
`used to protect against exonuclease degradation and 2’ sugar modification (2'-O-Me, 2'-F, others)
`is used for endonuclease resistance. With respect to maintenance of RNAi silencing activity,
`exonuclease-stabilizing modifications are all very well tolerated. Introduction of internal sugar
`modifications to protect against endonucleases is also generally tolerated but can be more
`dependent on the location of the modification within the duplex, with the sense strand being more
`
`

`

`DISCOVERY AND DEVELOPMENT OF RNAi THERAPEUTICS
`
`469
`
`amenable to modification than the antisense strand. Nevertheless, using simple, well-described
`modifications such as P = S, 2’-O-Me, and 2'-F,it is possible in most instances to fully nucle-
`ase-stabilize an siRNA duplex and maintain mRNAsilencing activity. Importantly, the degree of
`modifications required to fully stabilize the siRNA duplex can generally be limited in extent,
`thereby avoiding the toxicities associated with certain oligonucleotide chemistries.
`Improved nuclease stability is especially important in vive for siRNA duplexes that are exposed
`to nuclease-rich environments (such as blood) and are formulated using excipients that do not them-
`selves confer additional nuclease protection on the duplex. As might be expected in these situations,
`nuclease-stabilized siRNA show improved pharmacokinetic properties in vivo (Alnylam, unpublished
`results). In other situations, when delivering siRNA directly to more nuclease-amenable sites such
`as the lung or when delivering in conjunction with delivery agents such as liposomes, the degree of
`nuclease stabilization that is required can be reduced significantly. While the ability of an siRNA
`duplex to reach its target cell type intact is vitally important, whether nuclease protection confers a
`measurable benefit once an siRNA is inside the cell remains to be determined. While in vitro
`comparisons of naked siRNA versus fully stabilized siRNA do not reveal significant differences in
`longevity of mRNAsilencing [22], these studies have typically been performed using rapidly divid-
`ing cells, where dilution due to cell division, and not intracellular siRNAhalf-life governs the dura-
`tion of gene silencing [23]. With the recent advent of fluorescence resonance energy transfer studies
`using siRNA [24], it should be possible in the near future to understand more completely the intra-
`cellular benefit of nuclease stabilization on the longevity of RNAi-mediated silencing.
`
`16.2.4 Therapeutic Considerations
`
`With the identification of active siRNA, a set of rules were initially proposed for selecting
`potent siRNA duplex sequences [1]. Subsequently, a number of groups have developed more
`sophisticated algorithms based on empiric testing to identify multiple criteria that can contribute to
`defining an active siRNA [25-27]. Using current algorithms, sub-nM [C50 in vitro active siRNA
`can be routinely identified in a quarter to a half of the designed siRNA with a subset of siRNA often
`demonstrating low pM activity,
`In designing siRNAs for therapeutic purposes, other considerations beyond anactive target
`sequence exist. Wherepossible,it is desirable to identify target sequences that have identity across
`all the relevant species used in safety and efficacy studies, thus enabling developmentofa single
`drug candidate from research stage all the way through clinical trials. Other considerations in
`selecting a target sequence involve the presence of single-nucleotide polymorphisms and general
`ease of chemical synthesis.
`Predicting the nucleotide sequence and chemical modifications required to yield an ideal RNAi
`therapeutic still remains a work in progress. While much progress has béen madein understanding what
`attributes are required to identify an jn vitro active and stable siRNA, muchless is known about how
`well those attributestranslate into identifyingin vivo active siRNAs. For example, many ofthe issues
`around specificity are based on in vitro data and their in vivo' relevance remains to be determined.
`For example, the range of off-target genes identified in tissue culture can differ dramatically depend-
`ing upon the transfection method used to introduce siRNAsinto cells [28]. Likewise, induction of
`innate immune responses by certain siRNAs has beer shownto be cell-type specific [29]. At present,
`in order to identify robust in vitro active lead candidate siRNAs suitable for subsequent in vive study,
`the practical and prudent approachis to synthesize and screen a library of siRNA duplexes for potency,
`specificity, nuclease stability, and immunostimulatory activity. A good example of such an empiric
`approach was a screen that we conducted for siRNA targeting all vascular endothelial growth factor
`(VEGF)-A spliced isoforms to treat AMD. Over 200 siRNA with different sequences and chemistries
`were evaluated from which an optimized clinical candidate, ALN-VEGO1, was selected. This opti-
`mization procedure resulted in an siRNA with picomolarin vitro activity and sustained silencing in the
`relevant ocular cell type than was superior to other published VEGF siRNA compounds(Figure 16.3).
`
`

`

`470
`
`ANTISENSE DRUG TECHNOLOGY, SECOND EDITION
`
`(a)
`
`HeLa Celis
`
`0
`24h
`48h
`
`{
`I
`Transfect
`Change supernatant
`Quantitate VEGF
`siRNA
`by ELISA
`
`
`@ L2000
`
`mt ALN-VEGO1 MM
`
`=~ Luciferase siRNA
`
`=O=— Cand5 VEGF siRNA
`
`=O ALN-VEGO1
`
`
`
`%,VEGFprotein(rel.
`toL2000)
`
`100
`
`o Oo
`
`oO o
`
`40
`
`20
`
`0.11
`
`1
`
`10
`
`00
`
`.014
`
`siRNA (nM)
`
`(b)
`
`Human ARPE-19 Cells
`
`0
`1 day
`5 days
`10 days
`
`t
`;
`{
`|
`
`Transfect RPE
`>
`+r
`7%
`ae
`J
`with VEGF siRNA
`Change supernatant & quantitate VEGF by ELISA
`
`120
`
`3 = 100
`oo
`a.
`80
`LL
`oo 60
`wu”.
`= 2
`ae
`3
`
`40
`20
`
`0
`
`00.04 nM
`
`m30nM
`3.33 nM
`
`60.37 nM
`
`Figure 16.3 Identification of highly potent VEGF siRNA, ALN-VEGO1. HeLacells or ARPE-19 human retinal
`pigmentepithelial cell line'were plated in 96 well plates and transfected 24hlater with the indicated
`concentration of siRNA in Lipofectamine 2000. A Lipofectamine-alone control (L2000)is also indi-
`cated. At 24 h post-transfection, culture medium was completely removed and 100 il of fresh 10%
`FBS in DMEM added. Following this medium change, cultured supernatants from HeLa and
`ARPE-19 cells were collected 24 h (48 h post-transfection) and 96 h (5-day post-transfection)later,
`respectively. Fresh culture supernatant was added on day 5 to the confluent ARPE-19 cells and
`supernatantscollected again 5 days later (10-day post-transfection). Thus, the effect of siRNAinhi-
`bition on VEGF protein production was measured over different time periods post-transfection
`(HeLa, 24-48 h; ARPE-19, days 1-5, days 6-10). Quantitation of human VEGF protein in the cul-
`tured supernatants was by ELISA. Positive control is CandS hVEGF siRNA[32] and negative con-
`trols include an irrelevant siRNA (luciferase) and an ALN-VEGO1 siRNAcontaining four inverted
`nucleotide mismatches (ALN-VEGO1 MM).
`
`

`

`DISCOVERY AND DEVELOPMENT OF RNAi THERAPEUTICS
`
`471
`
`16.3.
`
`IN VIVO DELIVERY
`
`Effective delivery is the most challenging remaining consideration in the development of RNAi
`as a broad therapeutic platform. To date, animal studies using siRNA either have not employed
`additional formulation (i.e., “naked siRNA”) or have delivered siRNA formulated as conjugates,
`as liposome/lipoplexes, or as complexes (peptides, polymers, or antibodies), The route of admini-
`stration of siRNA has also ranged from local, direct delivery to systemic administration. Local
`delivery or “direct RNAi”has particular advantagesfor a developing technologyin that as with any
`pharmacologic approach, doses of siRNA required for efficacy are substantially lower when
`siRNA are injected into or administered at or near the targettissue. Direct delivery also allows for
`a more focused delivery of siRNA that might circumvent any theoretical undesired side effects
`resulting from systemic delivery. Systemic delivery of siRNA especially with cholesterol conju-
`gates and liposome formulations have also been widely explored with considerable success. While
`this section will provide a review of the different delivery approachesutilized with siRNA,it is not
`an exhaustive description ofall in vivo experimentation. Several recent publications offer such a
`review [20,30,31]}.
`
`16.3.1 Naked siRNA
`
`Manyreports describing success with RNAi in vivo involve direct delivery of “naked” siRNA to
`tissues such as eye, lung, and central nervous system, As used here, the term “naked” siRNArefers
`to the delivery of siRNA (unmodified or modified) in saline or other simple excipients such as 5%
`dextrose (DSW). The ease of formulation and administration using directdelivery of “naked” siRNA
`to tissues make this af attractive therapeutic approach. Not surprisingly, the initial development of
`RNAi therapeutics has focused on disease targets and clinical indications (AMD and RSVinfection)
`that allow for direct administration of siRNAto the diseased organ.
`
`16.3.1.1 Ocular
`
`Multiple examplesof efficacious local delivery of siRNA in the eye exist, where proof of con-
`cept has been attained in animal models of ocular neovascularization and scarring using both
`saline and lipid-based formulations [32-36]. Much evidence suggests that direct administration
`of “naked” siRNAis able to target cell types in the back of the eye and have profound disease-
`modifying effects, Using the optimized VEGF targeting siRNA ALN-VEGO1 described above,
`we have demonstrated robust specific inhibition of pathologic retinal neovascularization in a rat
`oxygen-induced model ofretinopathy (Figure 16.4). Following a single intravitreal injection of
`saline-formulated ALN-VEGO1, we achieved over 75% inhibition of pathological neovascular-
`ization with no effect on the normal retinal vasculature. The inhibition seen with ALN-VEGO1
`was both dose-dependent and specific as a mismatched siRNA showed no inhibition.
`Interestingly, the degree of inhibition seen with ALN-VEGO1 was dramatically more profound
`than that seen with either a VEGF,,.-specific aptamer (pegaptanib, approved for intravitreal use
`in AMD patients) or a VEGF-receptor immunoglobulin fusion protein (Figure 16.4). Separate ear-
`lier studies using lipid-formulated VEGF siRNA had shown a reduction of laser-induced
`choroidal neovascularization (CNV) in a mouse model of AMD [32]; this initial study was fol-
`lowed by nonhuman primate laser-induced CNV study where it was reported that intravitreal
`injection of a saline-formulated VEGF siRNA was well tolerated and efficacious [33], Lastly,
`intravitreal injection of saline-formulated siRNA targeting VEGF receptor-1 was effective in
`reducing the area of ocular neovascularization by 1/3 to 2/3 in two mouse models [34]. These
`encouraging proof-of-concept studies in animal models have lead toclinicaltrials of siRNA tar-
`geting the VEGF pathway in AMD,
`
`

`

`
`
`49°sinhibitionO18vsMM|
`
`(a)
`
`Nolnj)
`
`PBS
`
` siMM
`song
`
`(b)
`
`Irrelevant sIRNA
`
`
`Neovascularization(mm?)
`
`area(%) Noinj
`Normalvascular
`
`
`
`delivery of “naked” siRNAeither in saline or with excipients such as DSW orlung surfactants.
`
`472
`
`ANTISENSE DRUG TECHNOLOGY, SECOND EDITION
`
`(c)
`
`ae=
`3>a
`
`°=°
`
`p
`
`c= 8a
`
`S
`fs
`aw~
`
`o=
`
` siVEGF
`3g
`
` sIVEGF PegaptanibVEGF Re Ig
`GOpg
`16g
`tpg
`
`(d)
`
`ALN-VEGO1
`
`PBS
`
`siMM siVEGF siVEGFPegaptanib VEGF Re lg
`60u9
`3pg
`60ng
`16ug
`1g
`
`Figure 16.4 (See color insert following page 270.) ALN-VEGO01 specifically inhibits retinal neovasculariza-
`tion in a rat oxygen-induced retinopathy model.. Newborn rats were exposed to alternating high
`oxygen concentrations from days 0-14 as outlined previously [101]. On day 14, therapeutic
`agents were given onceintravitreally (5 pl volume) at the amounts indicatedand rats placed in
`room air for the following 6 days (days 14-20). On day 20, rats were sacrificed and flatmount reti-
`nal preparations stained with ADPase was used to quantitate (a) pathologic neovascularization
`and (b) normal vascular development; representative ADPaseflat mount preparations are shown
`following administration of(c) irrelevant control siRNA or (d) ALN-VEGO1. Experimental groups:
`no injection (No Inj), saline (PBS), high- and low-dose ALN-VEGO1 siRNA (siVEGF), high-dose
`ALN-VEGO01 mismatch siRNA (siMM), clinical-grade VEGF aptamer (Pegaptanib), and research-
`grade VEGF receptor immunoglobulin fusion protein from R&D Systems (VEGF Re Ig), All groups
`w