Intramural Coronary Delivery of AdvancedAntisense Oligonucleotides Reduces NeointimalFormation in the Porcine Stent Restenosis ModelNicholas N. Kipshidze, MD, PHD, FACC,* Han-Soo Kim, MD, FACC,† Patrick Iversen, PHD,‡Hamid A. Yazdi, MD,† Balram Bhargava, MD,† Gishel New, MD, FACC,*Roxana Mehran, MD, FACC,* Fermin Tio, MD,§ Christian Haudenschild, MD,�George Dangas, MD, PHD, FACC,* Gregg W. Stone, MD, FACC,* Sriram Iyer, MD, FACC,*Gary S. Roubin, MD, PHD, FACC,* Martin B. Leon, MD, FACC,* Jeffrey W. Moses, MD, FACC*New York, New York; Washington, DC; Portland, Oregon; San Antonio, Texas; and Rockville, Maryland

OBJECTIVES We evaluated the long-term influence of intramural delivery of advanced c-myc neutrallycharged antisense oligonucleotides (Resten-NG) on neointimal hyperplasia after stenting ina pig model.

BACKGROUND Neointimal hyperplasia after percutaneous coronary interventions is one of the key compo-nents of the restenotic process. The c-myc is a critical cell division cycle protein involved inthe formation of neointima.

METHODS In short-term experiments, different doses (from 500 �g to 5 mg) of Resten-NG or salinewere delivered to the stent implantation site with an infiltrator delivery system (InterventionalTechnologies, San Diego, California). Animals were euthanized at 2, 6 and 18 h afterinterventions, and excised vessels were analyzed for c-myc expression by Western blot. Inlong-term experiments, either saline or a dose of 1, 5 or 10 mg of Resten-NG was deliveredin the same fashion, and animals were euthanized at 28 days after the intervention.

RESULTS Western blot analysis demonstrated inhibition of c-myc expression and was dose dependent.Morphometry showed that the intimal area was 3.88 � 1.04 mm2 in the control. There wasstatistically significant reduction of intimal areas in the 5 and 10 mg groups (2.01 � 0.66 and1.95 � 0.91, respectively, p � 0.001) but no significant reduction in the 1 mg group (2.81 �0.56, p � 0.5) in comparison with control.

CONCLUSIONS This study demonstrated that intramural delivery of advanced c-myc neutrally chargedantisense morpholino compound completely inhibits c-myc expression and dramaticallyreduces neointimal formation in a dose dependent fashion in a porcine coronary stentrestenosis model, while allowing for complete vascular healing. (J Am Coll Cardiol 2002;39:1686–91) © 2002 by the American College of Cardiology Foundation

A potential application of molecular therapy with antisenseoligonucleotides is the prevention or treatment of restenosisafter coronary interventions (1,2). Balloon and/or stentinjuries result in neointimal hyperplasia that lead to recur-rent narrowing of the lumen within months after interven-tion (3). Alternatively, direct interference of the criticalsteps in the smooth muscle cell growth cycle has beenattempted, using antisense oligonucleotides in animal mod-els (4–7). Inhibition of several cellular proto-oncogeneshave been shown to inhibit smooth muscle cell proliferationin vitro and to reduce neointimal thickening in vivo (7–12).The clinical use of antisense technology, however, remainedlimited due to a relative lack of specificity, slow uptakeacross the cell membrane and rapid intracellular degradationof the oligonucleotide (13). Therefore, one study in humans

using local delivery c-myc antisense has yielded a negativeresult (14).

Recent studies have introduced phosphorodiamidatemorpholino oligomers (PMO), which represent an unusualDNA chemistry with a six-membered morpholino ringinstead of a deoxyribose sugar. In addition, the chargedphosphodiester and uncharged phosphorothioate linkagereplaces internucleotide linkage. The lack of internucleosidecharge allows the PMO to avoid the tendency of the morecommon phosphorothioate analogs to produce nonspecificeffects through binding of cellular and extracellular proteins(15–17). The PMO are resistant to the nucleases found inserum (18) and exhibit a high degree of specificity andefficacy in both in vitro and cell free translation studies(19–22).

It was also demonstrated that endoluminal delivery ofc-myc neutrally charged antisense oligonucleotide with atransport catheter reduces neointimal formation and posi-tively affects the vascular remodeling in the rabbit ballooninjury model (23).

The objective of the present study was: 1) to evaluate theeffect of local antisense therapy using advanced PMO withc-myc antisense oligonucleotides (Resten-NG) on c-myc

From the *Lenox Hill Heart and Vascular Institute and Cardiovascular ResearchFoundation, New York, New York; †Cardiology Research Institute, Washington,DC; ‡AVI BioPharma, Portland, Oregon; §Biomedical Research Foundation ofSouth Texas, San Antonio, Texas; and the �American Red Cross, Jerome HollandLaboratories, Rockville, Maryland. These studies were supported in part by a researchgrant from the Medical Research Fund at Lenox Hill Hospital (New York, NewYork), AVI BioPharma (Portland, Oregon) and Cook Cardiology (Broomfield,Colorado).

Manuscript received March 12, 2001; revised manuscript received February 21,2002, accepted February 27, 2002.

Journal of the American College of Cardiology Vol. 39, No. 10, 2002© 2002 by the American College of Cardiology Foundation ISSN 0735-1097/02/$22.00Published by Elsevier Science Inc. PII S0735-1097(02)01830-2

expression after stent implantation and 2) to determine thelong-term influence of local antisense therapy on neointimalhyperplasia after stenting in a pig coronary model.

METHODS

Oligonucleotide synthesis and purification. Phosphoro-diamidate morpholine oligomers were synthesized at AVIBioPharma as described previously (24). The Resten-NGsequence is complementary to the c-myc messenger ribo-nucleic acid at the translation initiation start site, 5� ACG-TTGAGGGGCATCGTCGC-3�, and the control se-quence employed in cell culture studies contains thesequence 5�-ACTGTGAGGGCGATCGCTGC-3�. ThePMOs were purified by ion exchange chromatography.Purity was �90% full length 20-mer as determined byreverse phase HPLC (high performance liquid chromatog-raphy) MALADI TOF mass spectroscopy.Western blot analysis. Western blot analysis involved a8.7 �g protein sample of HeLA cell lysate that was loadedonto a polyacrylamide gel, a 4% stacking gel and an 8%running gel at 100 V for 2 h. The gel was then electroblot-ted onto immobilon P membrane from Millipore (Bedford,Massachusetts). Actin served as an internal standard probedwith a mouse monoclonal primary antibody, AC40, fromSigma Chemical Co. (St. Louis, Missouri). The c-mycexpression was measured with a mouse monoclonal primaryantibody, C-33, from Santa Cruz Biotechnology (SantaCruz, California). Both actin and c-myc antibodies wereidentified with secondary goat anti-mouse IgG antibodiesconjugated with horse radish peroxidase from Santa CruzBiotechnology. The immunoblot bands were visualized bychemiluminescence with Amersham Pharmacia ECI detec-tion reagents (Piscataway, New Jersey).Animals and experimental protocol. Animals used in thisstudy received humane care in compliance with the “Prin-ciples of Laboratory Animal Care,” formulated by theNational Society for Medical Research, and the “Guide forthe Care and Use of Laboratory Animals,” prepared by theNational Academy of Sciences and published by the Na-tional Institutes of Health (NIH publication #85-23, re-vised 1985).

Eighteen female or male juvenile pigs fed on a standardnatural grain diet (25 to 30 kg) were sedated with acombination of ketamine (20 mg/kg) and xylazine (2 mg/kg) by intramuscular injection. The animals were givenpentobarbital (10 to 30 mg/kg IV) and were subsequentlyintubated and ventilated with oxygen (2 l/min) and isoflu-

rane 1% (1.5 l/min) using a respirator. Adequate anesthesiawas confirmed by the absence of a limb withdrawal reflex.Limb-lead electrocardiography and blood pressure (Honey-well E for M) were monitored throughout the procedure.

After placement of an 8F-introducer sheath in the rightcarotid artery by surgical cutdown, each animal receivedheparin (150 U/kg). Under fluoroscopic guidance, an 8Fguiding catheter was positioned in the left or right coronaryostium. Coronary angiography was performed after intra-coronary nitroglycerin (200 �g) administration and re-corded on cine film (Phillips Cardiodiagnost, Shelton,Connecticut).Antisense delivery. Antisense drug delivery was performedat the proximal and/or mid segments of the left anteriordescending, left circumflex or right coronary arteries via theInfiltrator Balloon Catheter (IBC) 3.0 to 3.5 mm indiameter and 15 mm in length (Interventional Technolo-gies, San Diego, California). The IBC was inflated to 2 atmsat the target site, and 0.4 ml of saline (control group) or 0.4 mlof antisense drug of different doses (from 0.5 to 10 mg) wereinfused intramurally via drug delivery InjectorPortsTM usingslow pressure hand delivery over approximately 20 s. Theballoon inflation pressure was then increased to 4 atms, and theIBC was held in place for an additional 30 s. The total timenecessary to deliver infusate was approximately 1 min for eachdelivery, and the balloon-to-artery ratio was kept at 1:1.

Coronary stenting was performed at the site of deliveryusing V-Flex stents 15 mm length (Cook Inc., Blooming-ton, Indiana), hand crimped on the balloon and deployed athigh pressure (10 to 14 atms � 30 sec). The stents weremounted on a balloon 3.5 to 4.0 mm in diameter and20 mm in length. The stent artery ratio was kept between1:1.1 and 1:1.2. Immediately post procedure, angiogramswere performed to assess vessel patency; the carotid sheathwas removed, the carotid artery was ligated, the skin wasclosed, and the animal was allowed to recover. All animalswere pre-treated with aspirin 325 mg and ticlopidine250 mg bid, from 24 h before the procedure until animalswere euthanized.C-myc expression after stent implantation and antisensedelivery. To evaluate the impact of antisense delivery onc-myc expression after stent implantation, five juvenile pigsweighing 30 to 35 kg underwent oversized multiple stentimplantation (four per animal) in the coronary artery. Inanother five animals (four treatment sites per animal), thiswas followed by local Resten-NG delivery (0.5, 1, 3 and5 mg), using the infiltrator catheter. In those animals, two ofthe sites were control sites (saline delivery). At either 2 to6 h, or 18 h, after the procedure, the pigs were euthanized,and treated tissue was analyzed by western blot for theexpression of c-myc.Long-term studies: influence on neointimal hyperplasia.Eight animals were treated in the same fashion. Treatmentsites were assigned at random to either control or treatmentwith three different doses of Resten-NG delivery (1, 5 and10 mg) with an IBC. At four weeks the animals were

Abbreviations and AcronymsH&E � hematoxylin and eosinIBC � Infiltrator Balloon CatheterPMO � phosphorothioate morpholino oligomersPTCA � percutaneous transluminal coronary angioplastySMC � smooth muscle cell

1687JACC Vol. 39, No. 10, 2002 Kipshidze et al.May 15, 2002:1686–91 Intramural Delivery of c-myc

euthanized. The arteries were perfusion-fixed, and theinjured segments located with the guidance of the coronaryangiograms were dissected free from the heart. After arteryremoval, the hearts were sectioned transaxially at 1-cmintervals, and sections were examined grossly for evidence ofmyocardial damage. The stented segments were processedfor plastic embedding, staining and histomorphometricanalysis of arteries (5 to 8 sections per vessel were averaged,expressed as mean value � SD) by use of publishedmethods (25). The grading scheme was used (25,26) todetermine the maturity of vascular repair (endotheliali-zation, intimal fibrin, inflammation, intimal SMC content,adventitial fibrosis).Transmission electron microscopy. Endothelial structurewas studied using transmission electron microscopy. Seg-ments of arteries were fixed in 3% glutaraldehyde for 24 hand then transferred to sodium cacdylate buffer. Dehydratedspecimens were embedded in Polybed 812 (Polysciences,Warrington, Pennsylvania), thin-sectioned and stained withuranyl acetate and lead citrate, and examined with a Philips301 electron microscope (Eindhoven, The Netherlands).Statistical evaluation. Data (mean � SD) were analyzedfor overall differences between treatment groups, usingone-way analysis of variance with the Bonferroni correction.Comparison of the mean values with a p value of �0.05 wasconsidered statistically different. All statistics were per-formed using SPSS 10.0 for Windows (SPSS, Inc., Chi-cago, Illinois). The intimal area and injury score werecorrelated using linear regression analysis.

RESULTS

Eighteen animals underwent intervention on 72 coronaryarteries. The antisense drug was delivered to segments ofeach coronary artery, and coronary stents were implantedsuccessfully at the sites where the antisense drug wasdelivered. All pigs survived to completion of the four-weekstudy without evidence of myocardial infarction on grossinspection and also after histological evaluation.C-myc expression after balloon angioplasty and antisensedelivery. High performance liquid chromatography analy-sis of vessel tissue for Resten-NG showed dose dependentdelivery of Resten-NG from 0.5 to 5 mg into the injuredsegment. Local delivery of Resten-NG into coronary arter-ies by infiltrator catheter blocked expression of c-mycinduced by stent implantation by western blot analysis ofvessel tissue and was dose-dependent (Fig. 1).Histological and morphometric analysis. All stents werewell developed within the vessel, resulting in thinning of themedia adjacent to the stent struts. In the rare vessels withstent protrusion into the adventitia, there was evidence ofperivascular hemorrhage (Fig. 2). No cases of thrombosis ofthe treated segment were observed in any of the treatmentgroups (Table 1). We observed complete healing withvirtually no toxicity in all treatment groups. Re-endothelialization was complete in all treatment groups

(Table 1). Transmission electron microscopy of antisense-treated vessels demonstrated normal appearance of endo-thelial cells with tight junctions between cells. There was nosignificant intimal fibrin in any groups. Inflammation scoreas well as adventitial fibrosis was similar in control andantisense treatment arteries (Table 1). The neointima fromantisense treated arteries with doses of 5 and 10 mg wassmaller in size than the controls. Control arteries exhibiteda substantial neointima consisting mostly of stellate andspindle-shaped cells in a loose extracellular matrix. In theantisense treated arteries, the cells of the neointima weremorphologically similar to controls.

The quantitative histomorphometry results of the controland antisense treated arteries are shown in Table 2. Nosignificant difference was seen in vessel size between controland antisense treated groups. These results demonstrate astatistically significant reduction of intimal area (IA) and IAnormalized to injury score (IS) after treatment with anti-sense of 5 and 10 mg. There was a strong trend favoring the

Figure 2. Cross-section samples of the control and the three dosage groupsshowing area percent stenosis of 69%, 45%, 32% and 25%, respectively(Metachromatic stain, Original magnification s.09X).

Figure 1. Data from Western blot analysis of porcine coronary arteries forc-myc and actin protein expression. Two hours following: (1) Stent; (2)Stent � 0.5 mg Resten-NG; (3) Stent � 1.0 mg Resten-NG; (4) Stent �3.0 mg Resten-NG; (5) Stent � 5.0 mg Resten-NG.

1688 Kipshidze et al. JACC Vol. 39, No. 10, 2002Intramural Delivery of c-myc May 15, 2002:1686–91

1-mg group versus control, and a significantly larger lumenarea with use of antisense treatment with 5 and 10 mg versuscontrol. The regression analysis between IS and IA dem-onstrated that the significant decrease in slope betweentreated and control groups indicates that local delivery ofantisense results in significantly less intimal growth despitesimilar degrees of arterial injury (Fig. 3).

DISCUSSION

Rationale for vascular applications of antisense molecu-lar therapy. Because of the proliferative component of therestenotic process, it is not surprising that most of theattention of antisense technology has focused on the short-lived regulators of the final common pathway of mitogenicstimuli, the cell cycle. Inhibition of the production of severalof the mediators of the cell cycle with antisense oligonucle-otides has been shown to be effective for the prevention ofrestenosis in several different animal models of vascularinjury (5–11).

However, the results of a single-center clinical trialperformed in Rotterdam that examined the effectiveness ofan antisense compound directed against c-myc were recentlyreported with disappointing results (14). The authors used aself-expanding stent, which can cause chronic injury tostented arteries. In these circumstances a single injection ofantisense may not be adequate to reduce a neointimalresponse to ongoing injury (vessel stretch). Additionally, the

RNAse H competent nucleotide chemistries, such as thephosphorothioate used in the Investigation by the Thorax-center on Antisense DNA Given by Local Delivery andAssessed by IVUS after Coronary Stenting (ITALICS)clinical trial, lacks specificity, in part because duplexes asshort as five base pairs in length can be cleaved by RNAseH (26). Cleavage of the RNA in such short duplexes mightoccur at a rate of 1 in every 1,000 bases of RNA sequence ornearly once in every transcript from every gene. Hence, theadvantages of the PMO chemistry represent fundamentalimprovements in the selective inhibition of target genemRNA over earlier antisense approaches and, therefore,worth consideration in the prevention of restenosis.

We choose early onco-gene c-myc as a target because ofits multiple role in development of neointimal hyperplasia.It was previously demonstrated that, along with the pro-found antiproliferative and antimigrative effects, c-mycantisense has significant anti-inflammatory properties, andmost importantly, it also inhibits matrix production (23).

Indeed, we recently evaluated the long-term impact oflocal delivery of c-myc neutrally charged antisense oligonu-cleotides on intimal hyperplasia after percutaneous translu-minal coronary angioplasty in a rabbit model (23). Mor-phometry at eight weeks showed significant inhibition ofintimal hyperplasia.Comparison with other antirestenotic strategies. Becauseof its low toxicity and local application, intramural deliveryof antisense may be theoretically more attractive than othermethods of suppressing excessive intimal response aftermechanical injury (radiation, etc.).

A large number of experimental and clinical studies havedemonstrated that endoluminal-ionizing radiation reducescellular proliferation and restenosis (27). However, edgeeffect, late thrombosis, and increased risk of myocardialinfarction represent significant limitations of this therapeu-tic modality (27,28).

Most recently, local antiproliferative strategies, includingpharmacological stent coatings (actinomycin-D, paclitaxel,rapamycin, etc.), have demonstrated inhibition of smoothmuscle cell proliferation in vitro, have reduced neointimalthickening in animal models of restenosis (26,29,30) andhave produced promising results in a pilot human study

Table 1. Antisense Histopathologic Findings

Tissue Event

Antisense Dose

0 �g 1 �g 5 �g 10 �g

Endothelializationscore

3� 3� 3� 3�

Inflammation score 0.47 � 0.49 0.29 � 0.25 0.39 � 0.47 0.25 � 0.17Intimal vascularity 0.47 � 0.45 0.08 � 0.13 0.33 � 0.60 0.13 � 0.32Medial wall tissue

necrosisND ND ND ND

Intimal fibrin 0.53 � 0.39 0.58 � 0.30 0.33 � 0.25 0.50 � 0.35Neointimal SMC 2.44 � 0.37 2.33 � 0.21 2.81 � 0.33 2.78 � 0.30Adventitial fibrosis 0.40 � 0.36 0.48 � 0.2 0.58 � 0.64 0.40 � 0.8Thrombosis ND ND ND ND

Data are presented as the mean value � SD. (Grading scheme, see Methods).ND � not detected; SMC � smooth muscle cells.

Table 2. Histomorphometric Results in Porcine Coronary Arteries 28 Days After Local Deliveryof Antisense and Stent Implantation

Control(n � 6)

1 mg(n � 8)

5 mg(n � 9)

10 mg(n � 7) p Value*

Vessel area (mm2) 9.29 � 1.24 9.54 � 1.79 9.02 � 1.54 9.72 � 1.98 0.262Stent strut area (mm2) 7.14 � 1.04 7.56 � 1.48 6.86 � 1.11 7.63 � 1.59 0.585Luminal area (mm2) 3.26 � 1.57 4.76 � 1.85 4.91 � 1.36 5.62 � 1.40† 0.062IA (mm2) 3.88 � 1.04 2.81 � 0.56 2.01 � 0.66† 1.95 � 0.91† �0.001Medial area (mm2) 2.05 � 0.31 1.97 � 0.42 2.04 � 0.42 2.16 � 0.49 0.459IS 0.95 � 0.46 0.91 � 0.46 0.90 � 0.34 0.94 � 0.44 0.546IA/IS 4.08 � 0.76 3.09 � 0.50 2.17 � 0.67† 2.13 � 0.55† �0.001

*By analysis of variance (ANOVA). †p � 0.05 vs. control (ANOVA post-test with Bonferroni correction). Data are presentedas the mean value � SD.

IA � intima area; IS � injury score.

1689JACC Vol. 39, No. 10, 2002 Kipshidze et al.May 15, 2002:1686–91 Intramural Delivery of c-myc

(31). However high local toxicity may alter the endotheli-alization process after stent implantation and may cause latecomplications.

By contrast with other chemotherapeutics (paclitaxel,actinomycin-D), Resten-NG inhibits cell cycle in the G-1.Compounds that inhibit the cell cycle in the early phase areoften less toxic. Therefore, Resten-NG as well as Rapamy-cin fit this description. Additionally, antisense can inhibitthe cell cycle at a number of points, so it may be effectiveregardless of the stage of cell growth at which the drug isapplied.

This study also showed that re-endothelialization wascompleted at 4 weeks in all treatment groups. Vascularhealing, inflammation, fibrin deposition and adventitialfibrosis was similar to control arteries.Study limitations. The major limitation of this study isthat relevance of different animal models of neointimalhyperplasia to human restenotic process is not yet wellestablished. The long-term effect of local delivery of anti-sense is unknown. However, observed complete vascularhealing at 28 days after antisense delivery and stent implan-

tation may suggest that possible late complications, includ-ing late thrombosis, are not likely to take place at three or sixmonths in this animal model or in clinical circumstances.

CONCLUSIONS

This study demonstrated that intramural delivery of ad-vanced c-myc neutrally charged antisense morpholino com-pound inhibits c-myc expression and dramatically reducesneointimal formation in a dose dependent fashion in aporcine coronary stent restenosis model while allowing forcomplete vascular healing. Clinical trials (AVI-4126) areunderway to further establish the efficacy of this interven-tion in man.

AcknowledgmentsThe authors are deeply indebted to John Petersen, MD, forhis review of the manuscript, Cathy Kennedy, for copyedit-ing and other editorial assistance, and Martin Fahy, biosta-tistician of the Cardiovascular Research Foundation, forstatistical analysis.

Figure 3. Regression analysis between injury score and intimal area. The significant decrease in slope between treated and control groups indicates that localdelivery of antisense results in significantly less intimal growth despite similar degrees of arterial injury. Each data point represents a single treated arterialsegment.

1690 Kipshidze et al. JACC Vol. 39, No. 10, 2002Intramural Delivery of c-myc May 15, 2002:1686–91

Reprint requests and correspondence: Dr. Nicholas Kipshidze,Lenox Hill Heart and Vascular Institute, 130 East 77th Street,Black Hall-9th Floor, New York, New York 10021. E-mail:nkipshidze@lenoxhill.net.

REFERENCES

1. Speir E, Epstein SE. Inhibition of smooth muscle cell proliferation byan antisense oligodeoxynucleotide targeting the messenger RNAencoding proliferating cell nuclear antigen. Circulation 1992;86:538–47.

2. Simons M, Rosenberg RD. Antisense nonmuscle myosin heavy chainand c-myb oligonucleotides suppress smooth muscle proliferation invitro. Circ Res 1992;70:835–43.

3. Topol EJ, Serruys PW. Frontiers in interventional cardiology. Circu-lation 1998;98:1802–20.

4. Simons M, Edelman ER, DeKeyser J-L, et al. Antisense c-myboligonucleotides inhibit arterial smooth muscle cell accumulation invivo. Nature 1992;359:67–70.

5. Morishita R, Gibbons GH, Ellison KE, et al. Single intraluminaldelivery of antisense cdc2 kinase and proliferating cell nuclear antigenoligonucleotides results in chronic inhibition of neointimal hyperpla-sia. Proc Natl Acad Sci U S A 1993;90:8474–8.

6. Zamecnik P, Stephenson M. Inhibition of Rous sarcoma virusreplication and cell transformation by a specific oligodeoxynucleotide.Proc Natl Acad Sci U S A 1978;75:280–4.

7. Morishita R, Gibbons GH, Ellison KE, et al. Intimal hyperplasia aftervascular injury in inhibited by antisense cdk 2 kinase oligonucleotides.J Clin Invest 1994;93:1458–64.

8. Edelman ER, Simons M, Sirois MG, et al. C-myc in vasculoprolif-erative disease. Circ Res 1995;76:176–82.

9. Shi Y, Fard A, Galen A, et al. Transcatheter delivery of c-mycantisense oligomers reduced neointimal formation in a porcine modelof coronary artery balloon injury. Circulation 1994;90:944–51.

10. Robinson KA, Chronos NAF, Schieffer E, et al. Endoluminal localdelivery of PCNA/cdc2 antisense oligonucleotides by porous ballooncatheter does not affect neointima formation or vessel size in the pigcoronary artery model of postangioplasty restenosis. Cathet CardiovascDiagn 1997;41:348–53.

11. Gunn J, Holt CM, Francis SE, et al. The effect of oligonucleotides toc-myb on vascular smooth muscle cell proliferation and neointimaformation after porcine coronary angioplasty. Circ Res 1997;80:520–31.

12. Ehsan A, Mann MJ, DellAcqua G, Dzau VJ. Long-term stabilizationof vein graft wall architecture and prolonged resistance to experimentalatherosclerosis after E2F decoy oligonucleotide gene therapy. J ThoracCardiovasc Surg 2001;121:714–22.

13. Zalewski A, Shi Y, Mannion JD, et al. Synthetic DNA-basedcompounds for the prevention of coronary restenosis. In: WickstromE, editor. Clinical Trials of Genetic Therapy With Antisense DNAand DNA Vectors. New York, NY: Marcell Dekker, 1998:363–93.

14. Kutryk MJ, Foley DP, van den Brand M, et al. Local intracoronaryadministration of antisense oligonucleotide against c-myc for theprevention of in-stent restenosis. Results of the randomized investi-

gation by the Thoracenter of antisense DNA using local delivery andIVUS after coronary stenting (ITALICS) trial. J Am Coll Cardiol2002;39:281–7.

15. Stein C, Cheng Y. Antisense oligonucleotideds as therapeutic agents:is the bullet really magical? Science 1993;261:1004–12.

16. Stein C. Does antisense exist? Nat Med 1995;1:1119–21.17. Shocman RL, Hartig R, Huang Y, et al. Fluorescence microscopic

comparison of the binding of phosphodiester and phosphothioateoligodeoxyribonucleotides to subcellular structures, including interme-diate filaments, the endoplasmic reticulum and the nuclear interior.Antisense Nucleic Acid Drug Dev 1997;7:291.

18. Hudziak RM, Barofsky E, Barofsky DF, et al. Resistance of morpho-lino phosphorodiamidate oligomers to enzymatic degradation. Anti-sense Nucleic Acid Drug Dev 1996;6:267–72.

19. Stein D, Foster E, Huang S, et al. A specificity comparison of fourantisense types: morpholino, 2�-O-methyl RNA, DNA, and phospho-rothioate DNA. Antisense Nucleic Acid Drug Dev 1997;7:151–7.

20. Summerton J, Stein D, Huang B, et al. Morpholino and phosphoro-thioate antisense oligomers compared in cell-free and in-cell systems.Antisense Nucleic Acid Drug Dev 1997;7:63–75.

21. Taylor MF, Paulaskis JD, Weller DD, et al. In vivo efficacy ofmorpholino modified oligomers directed against tumor necrosis factor-alpha mRNA. J Biol Chem 1996;271:17445–52.

22. Taylor MF, Paulaskis JD, Weller DD, et al. Comparison of efficacy ofantisense oligomers directed toward TNF-alpha in helper T andmacrophage cell lines. Cytokine 1997;9:672–81.

23. Kipshidze N, Keane E, Stein D, et al. Local delivery of c-myc neutrallycharged antisense oligonucleotides with transport catheter inhibitsmyointimal hyperplasia and positively affects vascular remodeling inthe rabbit balloon injury model. Catheter Cardiovasc Interv 2001;54:247–56.

24. Summerton J, Weller D. Morpholino antisense oligomers: design,preparation and properties. Antisense Nucleic Acid Drug Dev 1997;7:63–70.

25. Kornowski R, Hong MK, Tio FO, et al. In-stent restenosis: contri-butions of inflammatory responses and arterial injury to neointimalhyperplasia. J Am Coll Cardiol 1998;31:224–30.

26. Suzuki T, Kopia G, Hayashi S, et al. Stent-based delivery of sirolimusreduces neointimal formation in a porcine coronary model. Circulation2001;104:1188–93.

27. Larrouy B, Boiziau C, Toulme J, et al. RNAse H is responsible fornon-sequence specific inhibition of in vitro translation by 2�-O-alkylchimeric oligonucleotides: high affinity or selectivity, a dilemma todesign antisense oligomers. Nucleic Acids Res 1995;23:3434–40.

28. Leon MB, Teirstein PS, Moses JW, et al. Localized intracoronarygamma-radiation therapy to inhibit the recurrence of restenosis afterstenting. N Engl J Med 2001;344:250–6.

29. Herdeg C, Oberhoff M, Baumbach A, et al. Local paclitaxel deliveryfor the prevention of restenosis: biological effects and efficacy in vivo.J Am Coll Cardiol 2000;35:1969–76.

30. Farb A, Heller P, Shroff S, et al. Pathological analysis of local deliveryof paclitaxel via a polymer-coated stent. Circulation 2001;104:473–9.

31. Sousa JE, Costa MA, Abizaid A, et al. Lack of neointimal prolifera-tion after implantation of sirolimus-coated stents in human coronaryarteries: a quantitative coronary angiography and three-dimensionalintravascular ultrasound study. Circulation 2001;103:192–5.

1691JACC Vol. 39, No. 10, 2002 Kipshidze et al.May 15, 2002:1686–91 Intramural Delivery of c-myc