- Email: [email protected]
Developments in catheter technologies
JOURNAL OF VASCULARSURGERY
Lifeline Research Meeting Abstracts 1297
Volume31, Number6
against expressed tissue factor. In vitro and in vivo testing of this molecule suggests it retains the properties of both parents and has enhanced effect on retention of the glycoprotein I I b / I I I a receptor blocker compared with the receptor blocker alone. Methods Murine antirabbit G P I I b / I I I a receptor antibody (AZ-1) was conjugated with the antirabbit tissue factor antibody (AP-1) using chemical recombination of antibody fragments. The activity of this novel bispecific product was tested using platelet aggregometry and in a clotting assay designed to measure the inhibitory effects on platelet GPIIb/IIIa receptors and tissue factor respectively. In an ex vivo model the conjugate was delivered to rabbit aorta segments using a microporous infusion catheter (Cordis Corporation, US). The conjugate was mounted in a perfusion circuit to measure retention over 12 hours compared with the original anti-GPIIb/IIIa antibody. Results The conjugate inhibited adenosine diphosphate (ADP) induced platelet aggregation almost completely at a concentration of 45 btg per 108 platelets. It also prolonged tissue factor induced clotting time fourfold at a concentration of 6.5 btg/mL. The retention o f the conjugate was enhanced; 21.8% remained in the vessel segments after 12 hours compared with 10.6% in the control group (P < .05). Conclusion It is possible to construct a conjugate of antiG P I I b / I I I a and tissue factor antibodies that retain biological function and have together enhanced retention in the vessel wall. Ongoing in vivo studies and future trials will be presented. Our goal is to identify for any specific process a new antigen or exposure of previously hidden vessel wall component that will allow us to construct molecules that can be delivered systemically in less than normal systemic doses such that they target the pathology, the so called "magic bullet." Dealing with the problems of thrombolysis resistance, thrombolytic failure, and reocclusion are the current aims of our studies. DEVELOPMENTS
IN CATHETER
TECH-NOLOGIES
Steven Bailey, MD, FACC University of Texas Health Sciences Center at San Antonio San Antonio, Tex
Percutaneous coronary interventions, despite the use of endovascular stents, still have limitations due
to early thrombotic events as well as late neointimal proliferation. 1-4 Thrombus formation has been aggressively addressed using new oral and intravenous systemic antiplatelet therapy and improved flow with stcnts. S-7 Despite these newer therapies there are still major problems such as acute myocardial ischemia and saphenous vein graft degeneration. 8-1° Progressive disease in saphenous vein grafts is often with a friable and thrombotic base that continues to represent a challenging scenario for percutaneous interventions. Local delivery of pharmacologic agents is now possible using catheters specifically designed for regional or site specific delivery)l, ~2 These catheters have been developed to allow an increased volume of agent to be infused at a chosen site in an arterial segment. With the use of the properties of passive diffusion or active infusion, these catheters will deliver pharmaceuticals or genes into the arterial wall. Clinical approaches have used the ability of the new catheters to approach this problem in one of three techniques: (1) low-pressure diffusion, (2) low volume infusion, or (3) direct injection into the artery wall or pericardium. While each catheter system has its own unique characteristics, the current delivery systems have equivalent efficiencies of approximately 0.1% to 1.0%.13 While low, this still represents nearly a 100fold increase over that seen with systemic administration or infusion through a guide catheter. The limiting problem has not been identifying a delivery catheter but determining what pharmaceutical agent should be administered and what specific clinical problem we are treating. Coronary dissection and thrombus formation resulting in acute and subacute vessel closure remain significant problems. 14 The decision of which agents to use for local delivery centered on those agents with prior FDA approval. The widespread availability of thrombolytic and antiplatelet therapies resulted in the early use of these agents for local administration. The Dispatch catheter was one of the first systems evaluated in the clinical arena as a local administration device for the indication of coronary thrombus. A nondilating catheter that allows antegrade blood flow while the drug infusion was occurring it also offered the oppornmity for prolonged administration of agents without causing myocardial ischemia. This catheter was extensively characterized in humans by Camenzind et all5 in a novel and elegant manner. Using a continuous infusion of tracer that did not have myocardial uptake with rapid clearance (MAG-3), these authors were able to demonstrate that the actual site of administration
JOURNAL OF VASCULAR SURGERY
1298 Lifeline Research Meeting Abstracts
could be imaged and that the washout from the site could be calculated and standard models applied. The Dispatch catheter has been evaluated in a porcine model for the amount o f intramural deposition o f uroldnase when delivered by different techniques. There are small but significant differences in the delivery efficiency between the different methods o f administration. The clinical use o f these catheters has been reported in over 500 patients to date from registry studies. The earliest studies were those using the first generation catheters. The Infusasleeve registry trials and the local PAMI registries b o t h evaluated the use o f the Infusasleeve catheter in delivering heparin locally. The Local PAMI study evaluated heparin administered as an adjunct to direct angioplasty in the setting o f acute myocardial infarction. This study will be discussed in more detail below under the discussion o f using locally delivered heparin. Subsequent to this laboratory evaluation, the Dispatch catheter has been evaluated in two clinical trials. The first trial was a single center investigation o f uroldnase administration in 13 patients with vain graft disease by Glazier et al.17 Despite the use o f urokinase, two patients developed no fellow in patients with a large thrombus burden. The late clinical follow up was disappointing, with four o f the 12 patients returning with recurrence of clinical symptoms and a fifth patient experiencing late sudden death. The concept o f giving local administration o f urokinase was explored in more detail in the Dispatch and Uroldnase in the Elective treatment o f ghrombus [ D U E T ] trial. This randomized multicenter trial also evaluated the use o f the Dispatch catheter in saphenous vein grafts. It was initiated at the same time that intravenous I I b / I I I a antiplatelet therapies were being investigated. This resulted in a limited enrollment o f 53 patients. The trial was discontinued early because o f an inability to enroll patients as well as an interim analysis that demonstrated a trend toward more complications during the course o f the investigation. Patients who had the dispatch catheter used were more prone to early distal embolization and vascular complications. This study did not address the issue o f I I b / I I I a agent infusion through a local infusion catheter. We have subsequently examined the use ofabciximab using local delivery catheters 18 in a group o f patients with angiographically and angioscopically defined thrombus. The abciximab was delivered as 10 mg in 5 cc over 5 minutes. Angioscopy was performed within 5 minutes after abciximab infusion to
ju~e 2000
confirm the effects o f local abciximab infusion. Precedence for local delivery o f I I b / I I I a agents can be found in the publication by Aggarawal et al. 19 These investigators used a polymer-coated stent to which a mutine antibody to the I I b / I I I a locus was adsorbed. This system was demonstrated to decrease acute platelet deposition in a dose-dependent manner. While the amount o f the antibody was insufficient to inhibit 80% o f platelets it was sufficient to "passivate" the first layer ofplatelets that would have adhered to the stent surface. Unfortunately, this did not result in a reduction in late neointimal proliferation. It is entirely possible that a similar surface passivation can occur when I I b / I I I a agents are delivered using a local delivery catheter system. The time course o f elution has not been characterized for local delivery o f I I b / I I I a agents. Further support for the use o f I I b / I I I a agents is found in the study by Robinson et al.20 They used an external flow loop model to evaluate the usc o f I I b / I I I a agents bound to stents. T h e y demonstrated that there was a marked suppression o f platelet adhesion on stents where the I I b / I I I a agents were applied. Prior studies have also shown that other coatings such as heparin and phosphytidylcholine have decreased both acute platelet deposition and late intimal proliferation.21-24 The administration o f uroldnase has also been investigated in a single-center investigation by Glazier et al. 2s This study evaluated the use o f a hydrogel-coated balloon onto which uroldnase had been adsorbed. The investigators immersed the balloon into a solution o f uroldnase containing 50,000 U / c c . Ninety-five patients were enrolled because o f the presence o f thrombus angiographically (acute MI = 50; postinfarction angina, n = 23; and unstable angina n = 22). While no deaths or emergency bypass procedures occurred, there was a 7.4% incidence o f acute complications with no reflow in three patients, distal embolization in one patient, and late vessel closure in one patient. The late clinical event rate compose d o f death, myocardial infarction, and recurrent angina was 30.5%. This study may represent the fact that thrombolytic therapy increases platelet activation in the setting o f acute ischemia and therefore may increase the risk o f embolization and late platelet driven thrombotic events. This same group 26 used the Dispatch catheter in a small pilot series o f 13 patients with saphenous vein grafts whose average age was 11 years. This population all had thrombosis or occlusion o f aged saphenous, and 12 o f the 13 patients presented with unstable angina pectoris. These patients underwent
JOURNAL OF VASCULAR SURGERY
Volume31, Number 6
infusion o f uroldnase t h r o u g h a dispatch catheter followed by definitive therapy with a balloon inflation or stent placement. The dispatch catheter is a nondilating catheter, and the residual stenosis after uroldnase infusion was still 50% _+ 16%. After definitive therapy, the minimal lesion stenosis fell to 26% _+ 15%. Procedural success was 14 of 15 lesions (12 patients). Two patients had transient no reflow with one patient evolving a n o n - q wave myocardial infarction. Late clinical events were present in four patients with restenosis. A fifth patient with severe three vessel coronary disease had sudden death at home not felt to be related to the therapy. Local delivery ofunfractionated and fractionated heparin has been evaluated. This small molecule has shown promise in prior animal studies in reducing restenosis. Prior animal models including rat, rabbit, and pig 27-3° have suggested that high dose systemic heparin could reduce late neointimal hyperplasia. The use o f low molccular weight heparin was evaluated in the animal model by H o n g et al.31 This study used the channeled balloon to deliver enoxaparin (10 m g / k g ) in a rabbit iliac model. This study demonstrated that acute administration o f a single dose o f low molecular weight heparin was no more effective than control. The combination o f acute administration followed by systemic administration was effective in decreasing late neointimal proliferation. Efficacy o f delivery was only 0.1% to 0.2%. Measurement o f systemic anti-Xa levels demonstrated that the majority o f the enoxaparin was distributed systemically. It is unclear whether the low rate o f local delivery may explain the lack o f efficacy or it is simply not an effective agent for neointimal hyperplasia. It is also unclear if the additional injury from the balloon delivery played a role in the late outcomes. Finally, neointimal hyperplasia has been demonstrated to be a predominate mechanism o f restenosis after stcnt placcment as contrasted with vessel remodeling that occurs after balloon angioplasty, a2 Systemic therapy with nadroparin, a low molecular weight heparin, was attempted in 354 patients in the Fraxiparine Angioplastie Coronaire Transluminale (FACT) trialY There was no difference in either clinical or angiographic outcomes at the 6-month end point. This hypothesis has been tested by Kiesz et at 34 in the Polish Local delivery in the Nir stent trial ( P O L O N I A trial). This trial used low molecular weight heparin (enoxaparin) as a single dose infused into the artery using the transport balloon prior to stent placement. Both the acute outcomes as well as
Lifeline Research Meeting Abstracts 1299
the late clinical and angiographic outcomes were assessed. This trial demonstrated that use o f locally delivered enoxapafin was sufficient to avoid the need for systemic heparinization during the procedure. These data have been presented in preliminary form and demonstrate a significant decrease in neoinfimal hyperplasia as well as the late loss index. These data are the first study to demonstrate angiographic and clinical success using locally delivery in clinical cardiology. There was a publication by Oberhoff et al a5 on a second small series o f patients, in which the results o f local delivery o f intracoronary reviparin in 18 patients were reviewed. The primary success was 100% for delivery o f 7000 units o f intracoronary bolus followed by 1500 units intravenously. These patients also received subcutaneous reviparin 7000 units daily for the next 28 days. The angiographic restenosis rate from this registry was 28% ( 5 / 1 8 ) at 6 months in vessels that were 2.5 to 3.0 mm in diameter. This demonstrates the safety o f this technique in small vessels. It remains to be proved if this technique will reduce restenosis in a randomized trial. Unfractionated heparin has also been evaluated in animal models as a method o f decreasing neointimal hyperplasia. This technique has been tested in several animal models as well as with several delivery catheters. It is unclear what amount ofheparin must be available from local delivery. The only available study with quantitative information is from Mitchel et al. 36 This study demonstrated that the delivery o f only 0.5 to 0.6 units was sufficient to decrease early platelet deposition, but it took 30 to 60 units to attenuate smooth muscle cell proliferation. Clinical studies using heparin therapy delivered using catheter-based systems have been promising. LopezSandon et al37 demonstrated decreased neointimal proliferation at 28 days in a porcine model after infusion using a double balloon catheter. Heparin has been demonstrated to diffuse across the entire arterial wall and into the adventitia. The rate o f platelet deposition varies with the vessel injured. There is a threefold difference in platelet deposition between peripheral vessels and coronary vessels after balloon injury. The hypothesis that limiting acute platelet deposition will decrease late restenosis was the basis for the HIPS trial. This trial randomized 2 5 0 patients between 5000 units o f intracoronary heparin and intraluminal hepafin. The end points were angiographic stenosis as well as intravascular determined plaque volume. This study demonstrated that it was safe to administer heparin in this manner prior to stent implantation. There was no difference in late angiographic or clinical
JOURNAL OF VASCULAR SURGERY
1300 Lifeline Research Meeting Abstracts
outcome in this study. There were no differences in the demographics of the enrolled paticnts between the two groups of intramural versus intraluminal. The ages were the same (57.8 _+ 11.3 vs 59.6 _+10.5 years), male gender was similar (80.2% vs 75.0%) as was the incidence of unstable angina (75.8% vs 73.9%) and prior myocardial infarction (47.3% vs 39.1%). In addition the LVEFs were similar at 54.1% _+9.I% vs 54.3% _+8.1%. This procedure was very safe with a similar percentage o f procedural success o f 99.4% versus 100%. There were no differences in the late clinical outcomes with respect to revascularization. One patient suffered a q wave myocardial infarction and subsequently died in the intramural group, and the incidence o f non-q myocardial infarctions was no different between the groups (8.8% vs 12.5%). Note that the 6-month revascularization rate is exceptionally low in both groups. This is consistent with the large vessels that were entered into this study. The angiographic restcnosis rate was 12.7% in both groups, and there was no difference in late clinical outcomes. A primary end point of the study was intravascular ultrasound determined plaque volume inside the stented segment o f vessel. The intravascular ultrasound-determined vessel volume was 36.4 _+ 20.1 m L 3 in the intramural-treated group and 44.1 _+ 25.1 m L 3 in the intraluminal-control group. The limitations o f the HIPS trial included the use o f a low dose o f heparin for local administration. It is possible that the study dosage was inadequate to achieve sufficient local therapeutic levels. Animal trials had used equivalent doses 10 times higher than those used in this study. Alternatively, this form o f therapy may not be effective in preventing late intimal hyperplasia. This finding was predicted by the studies of Edelman et al38 who demonstrated that diffusion was not effective in treating the thick arteries found in diseased coronary vessels. They showed that adventitial delivery ofheparin was required for suppression ofneointimal proliferation. The use o f unfractionated heparin covalently bound to stents was evaluated in the heparin-coated stent study "Benestent II. "39 This trial evaluated a novel, covalently bound heparin-coated stent compared with balloon angioplasty. The subacute rate was exceptionally low; however, the rate o f restenosis in the stent group was not different than that predicted from the n o n c o a t e d stent used in the B E N E S T E N T I trial. This suggests that there is no antiproliferative effect o f the heparin coating. This mirrors data from animal studies that also failed to demonstrate a decrease in neointimal proliferation in
June 2000
a porcine model. 40 It is unclear whether higher doses o f heparin or noncovalently b o u n d heparin would have decreased the incidence o f neointimal proliferation. Direct thrombin inhibitors have also been used in an attempt to decrease late neointimal proliferation. The trial by Burchenal et al41 o f systemically delivered Bivalirudin (Hirulog) demonstrated that this agent was not effective in decreasing late neointimal proliferation compared with heparin therapy alone in a balloon angioplasty population. The therapy had only a 24-hour duration and is a significant limitation in study design as several studies have shown that thrombin generation c o m m o n l y occurs up to 48 hours after the procedure. This is similar to the data from I I b / I I I a agent's (Capture) trial42 that did not demonstrate a clinical benefit beyond 30 days when only a bolus and one hour infusion ofabciximab were administered after percutaneous interventions. Local delivery o f I I b / I I I a agents is being considered. The exciting data that clinical events are decreased up to 3 years after systemic administration are encouraging. T h e r e are anecdotal reports o f angiographic controlled studies that have demonstrated that single bolus administration o f abciximab decreases restenosis following percutaneous transluminal angioplasty in high risk patients. While this finding is contrary to the data that platelet deactivation requires suppression o f over 80% o f platelets, it may support the fact that restenosis may be more closely tied to the a 3 / b v receptor. Alternatively, "local platelet suppression" at the lesion site may be sufficient to allow adequate platelet inactivation to occur to limit local platelet deposits after an initial layer o f platelets are deposited. This basal layer o f platelets is important in protecting the vessel surface from further events. There are data that support the use o f locally delivered I I b / I I I a agents using stents. The work ofAggarwal et al43 that demonstrated that platelet deposition could be significantly inhibited by a murine antibody to the I I b / I I I a locus is an important finding. The authors found that using polymer-coated stents onto which the AZ1 antibody had been adsorbed inhibited platelet deposition both in vitro and in vivo. The elution curves demonstrated a biexponential curve with an early rapid phase followed by a slow elution such that 40% o f the AZ1 antibody was still present at 14 days. Platelet deposition as assessed by actual platelet counts and physiologic parameters such as cyclic flow variation were improved in the coated stent. Unfortunately, this study did not demonstrate any difference in the late intimal thickness or the
JOURNAL OF VASCULARSURGERY Volume 31, Number 6
intima to media thickness ratio in the A Z l - c o a t e d stent compared with the n o n - A Z T - c o a t e d stent. This finding may represent the fact that the stent had an inadequate a m o u n t o f the I I b / I I I a agent present or more likely the fact that a different receptor blockade may be required. The ability to use locally administered I I b / I I I a agents as single dose therapy for rapid dissolution o f thrombus has been demonstrated in our laboratory. We used angioscopy to evaluate lesions in patients with unstable angina and postinfarction angina to determine the presence ofthrombus. Abciximab was delivered through local delivery catheters such as the Infusasleeve (Local Med, Palo Alto, Calif) or the Dispatch catheter (Scimed). With the use o f only 10 mg o f abciximab, near complete resolution o f thrombus could be demonstrated within 10 minutes using the angioscope. The patient population was too small to allow any information regarding restenosis although only one o f the 11 patients returned with clinical restenosis. Clinical information regarding the use o f other agents such as Taxol is not yet available. This agent does have significant promise in the treatment o f neointimal proliferation. This agent acts in a different manner than agents that decrease cell proliferation. Taxol and other taxenes interfere with the ability o f microtubules. This interferes with the ability o f smooth muscle cells to change shape and move into the artery wall. This agent has been demonstrated by Sollott et al44 to decrease neointimal hyperplasia in a rat model at 11 days in a dose-dependent manner. Subsequent studies have been presented in abstract form that also indicate that this agent can be delivered through stents in a local manner to decrease late neointimal hyperplasia. This therapy has many similarities to the effects o f irradiation and may mimic the cytostatic rather than cytotoxic agents that have previously been used. Unfortunately, restenosis may not be significantly impacted if vessel recoil is not affected by this therapy. Axel et al45 have evaluated the use o f Taxol for local infusion using the microporous infusion catheter from Cordis corporation. This agent had both an in vitro effect as well as a significant effect in a thermal injury rabbit carotid model. The decrease in intimal hyperplasia was significantly greater in the treated animals than in the control animals. The neointimal area was 0.21 + m m 2 vs 0.36 + 0.29 m m 2 in the treated versus the control vessels. Importantly, this decrease was not accompanied by any delay in endothelialization. This differs from the findings seen after irradiation therapy in which the
Lifeline Research Meeting Abstracts 1 3 0 1
endothelial cell was significantly delayed. In addition the concentrations required to inhibit micrombule cross-linking were lO-fold lower than that required to inhibit cell proliferation. These levels are achievable using local delivery techniques. In addition the use o f paclitaxel combined with low dose locally delivery irradiation also appears to have promise as a new therapeutic modality. This technique could potentiate both the microtubular effect as well as the cytotoxic effect o f both therapies. Conclusions Local drug delivery o f thrombolytic agents such as urokinase has been shown to be safe and effective for the treatment o f intracoronary thrombus. The delivery o f GP I I b / I I I a agents is currently being evaluated in the same setting. Early experience suggests that local delivery o f these agents may be as effective as systemic delivery. W h e t h e r long-term benefits will be seen is unclear at this time. Current techniques o f local delivery using conventional doses o f heparin have not been shown to decrease restenosis in a prospective randomized trial or to decrease the amount ofneointimal hyperplasia. There is evidence that new therapies using cell cycle specific agents such as paclitaxel may inhibit cell proliferation and the production o f cxtracellular matrix. Current experience indicates that the choice o f the delivery system is important to the procedural outcomes as seen in the D U E T trial. Careful consideration will need to be given prior to future studies in characterizing the limitations o f delivery system and efficiency prior to embarking on future studies o f local therapy to decrease neointimal proliferation. REFERENCES 1, Hoffmann R, Mintz GS, Dussaillant GR, Popma JJ, Pichard AD, Sailer LF, et al. Patterns and mechanisms of in-stent restenosis: a serial intravascular ultrasound study. Circulation 1996;94:1247-54. 2. Faxon DP, Coats W, Currier J. Remodeling of the coronary artery after vascular injury. Prog Cardiovasc Dis 1997; 40:129-40. 3. Kimura T, Kaburagi S, Tamura T, Yokoi H, Nakagawa Y, Yokoi H, et al. Remodeling of human coronary arteries undergoing coronary angioplasty or atherectomy. Circulation 1997;96:475-83. 4. Kronowski R, Mintz GS, Kent KM, Pichard AD, Sailer LF, Bucher TA, et al. Increased restenosis in diabetes mellitus after coronary interventions is due to exaggerated intimal hyperplasia. Circulation 1997;95:1366-9. 5. Lefkovits J, Ivanhoe RJ, Califf RM, et al. Effects of platelet glycoprotein I I b / I I I a receptor blockade by a chimeric monoclol~al antibody (abciximab) on acute and six-month out-
1302
6.
7.
8. 9.
10.
11. 12. 13. 14.
15.
16.
17.
18.
19. 20. 21.
22.
IOURNAL OF VASCULARSURGERY lune 2000
Lifeline Research Meeting Abstracts
comes after percutaneous transluminal coronary angioplasty for acute myocardial infarction. Epic Investigators Am Cardiol 1996:77:1045-51. Tcheng JE. Glycoprotein I I b / I I I a receptor inhibitors: putting the EPIC, Impact II, Restore and Epilog trials into perspective. Am J Cardiol 1996;78:35-40. Topoi EJ. Prevention of cardiovascular ischennc complications with new platelet glycoprotein IIb/IIIa inhibitors. Am Heart i 1995;130:666-72. Banters C, Isner JM. The biology of restenosis. Prog Cardiovasc Dis 1997;40:107-16. Mintz GS, Popma II, Pichard AD, Kent KM, Satler LF, Wong SC, et al. Arterial remodeling after coronary angioplasty. Circulation 1996;94:35-43. Rogers C, Edelman ER. Endovascular stent design dictates experimental restenosis and flarombosis. Circulation I995; 91:29953001. Bailey SR. Local drug delivery: current applicatioris. Prog Cardiovasc Dis 1997;40:183-204. Brieger D, Topol E. Local drug delivery systems and prevention ofrestenosis. Cardiovasc Res 1997;35:405-13. Edelman ER, Lovich M. Drug delivery models transported to a new level. Nature Biotechnol 1998;16:136-7. Sehieman G, Cohen BM, Kozina l, Erickson JS, Podoloin RA, Peterson KL, et al. Intracoronary urokinase for intracoronary thrombus accumulation complicating percutaneous intraeoronary thrombus accumulation complicating percutaneous transluminal coronary angioplasty in acute in acute ischemic syndromes. Circulation 1990;82:2052-60. Camenz~fid E, Bakker A, Reijs A, Vangeijlswijk IM, Boersma E, Kutryk MJB, et al. Site-specific intracoronary heparin delivery in humans after balloon angioplasty: a radioisotopic assessment of regional pharmacokinetics. Circulation 1997; 96:154-65. Mitchet JF, Shwedick M, Alberghini TA, Iqa-fibbs D, McK~y RG. Catheter-based local thrombolysis with uro!dnase: comparative efficacy of intraluminal clot lysis with conventional urokinase infusion techniques in an in vivo porcine thrombus model. Cathet Cardiovasc Diagn 1997;41:293-302. Glazier JJ, Keman FJ, Bauer HI-I, Gram DB, Primiano, CA, Mitchel JF, et al. Treatment of thrombotic saphenous vein bypass grafts using local uroldnase infusion therapy with the dispatch catheter. Cathet Cardiovasc Diagn 1997;41:261-7. Bailey SR, O'Leary E, Chitton R. Angioscopic evaluation of site-specific administration of ReoPro. Cathet Cardiovasc Diagn 1997;42:181-4. Aggarwal RK, Ireland DC, Azrin MA, Ezekowitz MD, de Bono DP, Gershlick AH. Circulation 1996;94;3311-8. Robinson et al article on IIb/IIIa and flow loop of stents. DeScheerder IK, Wilczek KL, Verbeken EV, Vandorpe J, Lan PN, Schacht E, et al. Biocompatability of polymer-coated oversized metallic stents implanted in normal porcine coronary arteries. Elsevier Science Atherosclerosis 1995;114:105-14. Hubbell JA. Biomatetials in tissue engineering. Biotecfmology
1995;13:565-76. 23. Mckenna CJ, Camrud AR, Sangiorgi G, Kwon HM, Edwards WD, Holmes DR Jr, et al. Fibrin-film stenting in a porcine coronary injury model: efficacy and safety compared with uncoated Stents. J Am Coil Cardiol 1998;31:1434-8. 24. Lincoff AM, Furst JG, Ellis SG, Turch RJ, Topoi EJ. Sustained local delivery of dexamethasone by a novel intravascular eluting stent to prevent restenosis in the porcine coronary injury model. J Am Coil Cardiol 1997;29:808-16. 25. Glazier JJ, Hirst JA, Keman FJ, Fram DB, Eldin AM,
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38. 39.
40.
Primiano CA, et al. Site-specific intracoronary thrombolysis with urokinase-coated hydrogel balloons: acute and followup studies in 95 patients. Cathet Cardiovasc Diagn 1997; 41:246-53. Glazier Jl, Kernan FJ, Bauer FIH, Fram DB, Primiano CA, Mitchel JF, et al. Treatment of thrombotic saphenous Vein bypass grafts using local urokinase infusion therapy with the dispatch catheter. Cathet Cardiovasc Diagn 1997;41:261-7. Guyton JR, Rosneber RD, Clowes AW, Karnowsky MJ. Inhibition of rat arterial smooth muscle cell proliferation by hepatin. Circ Res 1980;46:625-34. Clowes AW, Clowes MM. Kinetics of cellular proliferation after arterial injury, II: inhibition of smooth muscle growth by heparin. Lab Invest 1985;52:611-16. Wolinsky H, Thung SN. Use of a perforated catheter to deliver concentrated heparin into the wall of the normal into the wall of the normal coronary artery. J Am Coll Cardiol 1990;15:475-81. Azrin MA, Mitchel IF, Fram DB, Pedersen CA, Cartun RW, Barry JJ, et al. Decreased platelet deposition and smooth muscle cell proliferation following intramural heparin delivery with Hydrogel-coated balloons. Circulation 1994;90: 433-41. Hong MK, Wong SC, Barry Jl, Bramwell O, Tjurmin A, Leon MB. Feasibility and efficacy of locally delivered enoxaparin via the channeled balloon catheter on smooth muscle cell proliferation following balloon injury in rabbits. Cathet Cardiovasc Diagn 1997;41:232-40. Hoffmann R, Mintz GS, Dussaillant Gtl, Popma JJ, Pichard AD, Satler LF, et al. Patterns and mechanisms of in-stent restenosis: a serial intravascular ultrasound study. Circulation I996;94:1247-54. Lablance JM, McFadden EP, Meneveau N, Lusson JR, Bertrand B, Metzger IP, et al. Effect of dadroparin, a lowmolecular weight heparin on clinical and angiographic restenosis after coronary balloon angioplasty: The FACT Study. Circulation 1997;96:3396-402. Kiesz RS, Buszman P, Martin JL, Deutsch E, Rozek NIM, Rweieki M, et al. Polish-American Local Lovenox Nir Stent Assessment Study (POLONIA): final results [abstract]. J Am Coll Cardiol 1999;33 Suppl A:4A. Oberhoff M, Banmbach A, Hermann T, Diehl C, Maier R, Athanasiadis A, et al. Local and systemic delivery of low molecular weight heparin following PTCA: acute results and 6-month follow-up of the initial experience with the porous balloon (PILOT-study). Preliminary Investigation of Local Therapy Using Porous PTCA Balloons. Cathet Cardiovasc Diagn 1998;44:267-74. Mitchel JF, Azrin MA, Fram DB, Lanrine MB, McKay RG. Localized delivery of heparin to angioplasty sites with iontophoresis. Cathet Cardiovasc Diagn 1997;41:315-23. Lopez-Sandon J, Sobrino N, Bamallo C, Lorenzo A, Jimenez l, Calvo L, et al. Locally delivered heparin reduces intimal hyperplasia and lumen stenosis following balloon injury in swine [abstract]. Eur Hear J Suppl 1993;14:191. Edelman ER, Lovich M. Drug delivery models transported to a new level. Nat Biotechnol 1998;16:136-7. Serruys P, Van Hour B, Bonnier H, Legrand V, Garcia V, Macaya C, et al. Randomized comparison of implantation of heparin-coated stents with balloon angioplasty in selected patients with coronary artery disease (Benestent II). Lancet 1998;352:673-81. Hardhammar PA, van Beusekom I-IM, Emanuelsson HU, Hoffman SH, Albertson PA, Verdouw PD, et al. Reduction
JOURNAL OF VASCULARSURGERY
Volume31, Number6
41.
42.
43.
44.
45.
in thrombotic events with heparin-coated Palmaz-Schatz stents in normal porcine coronary arteries. Circulation 1996;93:423-30. Burchenal JEB, Marks DS, Mann JT, Schweiger MJ, Rothman MT, Ganz P, et al. Effect of direct thrombin inhibition with Bivalirudin (Hirulog) on restenosis after coronary angioplasty. Am J Cardiol 1998;82:511-5. van de WerfF. More evidence for a beneficial effect ofplatelet glycoprotein IIb/IIIa-blockade during coronary interventions: latest results from the EPILOG and CAPTURE trials. Circulation 1996;93:637. Aggarwal RK, Ireland DC, Azrin MA, Ezekowitz MD, deBono DP, Gershlick Ah. Antithrombotic potential of polymer-coated stents eluting platelet glycoprotein I I b / I I I a receptor antibody. Circulation 1996;94:3311-7. Sollott SJ, Cheng L, Pauly RR, Jenkins GM, Monticone RE, Kuzuya M, et aft. Taxol inhibits neointimal smooth muscle cell accumulation after angioplasty in the rat. J Clin Invest 1995;95:1869-76. Axel DI, Kunert W, Goggelmann C, OberhoffM, Herdeg C, Kuttner A, et al. Paclitaxel inhibits arterial smooth muscle cell proliferation and migration in vitro and in vivo using local drug delivery. Circulation 1997;96:636-45.
FIBRIN-BASED MATRICES F O R A N G I O GENIC STIMULATION Andreas H. Zisch, MD U. Schenk B. Schwe#zer S. Sakiyama-Elbert J. C. Schense jr. A. Hubbell Inst#ute for Biomedical Engineering Zurich, Switzerland
Our laboratory's research in vascular tissue engineering focuses on two distinct goals: first, in the context of building blood vessels dc novo, we aim to engineer biomaterial surfaces that are able to attract and retain endothelial cells organized as two-dimensional endothelial cell sheet under flow conditions, similar to the the native luminal vessel wall. Second, featuring the three-dimensional organization of endothelial cells, the generation of biomaterial matrices, which when grafted to the wounded or ischemic tissue, locally stimulate neovessel formation in the surrounding of the graft and also in the graft itself. In both cases, the material designs aim the vascular grafts to become populated in situ with endothelial cells derived from the body's endogenous sources (ie, from the artery adjacent to the graft, from existing blood vessels, or from circulating endothelial progenitor cells). For that reason, these grafts must have properties specifically attractive for endothelial cells. Through basic research, a better knowledge has been developed about the molecular aspects and requirements for reendothe-
Lifeline Research Meeting Abstracts 1303
lialization and blood vessel formation. This biological information can be inferred on a biomaterial by incorporating peptide or protein motifs that are necessary for the recruitment and the binding of endothelial cell binding. Polypeptide growth factors, in particular vascular endothelial growth factor (VEGF), have evolved as critical for many aspects of endothelial cell biology, such as the maturation, the proliferation, the assembly, and the survival of endothelial cells. The Hubbell laboratory has recently developed methodology that allows fibrin to be modified by covalent incorporation of bioactive peptides or proteins during the coagulation process. The base matrix fibrin is a natural ingrowth matrix for endothelial cells. To confer the angiogenic features of VEGF upon fibrin, we used this methodology to covalently incorporate VEGF121, an otherwise diffusible VEGF isoform into the fibrin matrix. This VEGF121/fibrin matrix exhibits strong angiogenie responses both in vitro and in vivo as described below. Further, we will discuss a sophisticated and elegant fibrin modification scheme in which growth factors, such as the VEGF165 isoform, can be entrapped and retained within the matrix through their heparin-binding abilities. From these depots, growth factors can be released by controllable diffusion as well as the activity of matrix-degrading enzymes secreted by cells invading the matrix. The clinical demand for angiogenic materials-modified fibrin as material for clinics Design of materials that induce angiogenesis in vivo is important for many areas of tissue regeneration and wound healing. One of the more relevant and therefore well studied needs is in treatment of chronic wounds, such as diabetic ulcers, pressure ulcers, or venous static ulcers that are caused by deficient vascular supply. Scores of trials with ulcer patients have been conducted over the past years using growth factor proreins or cytokines as therapeutics, with many discouraging results. One likely explanation for the failure of many therapies is the current modes of growth factor delivery in the context of creams and sprays, which are often inappropriate for providing sustained delivery of active factors. Without enbedment, however, the soluble protein factors are cleared very rapidly from the target site, or never reach the target in sufficient quanrifles, or reach it only at extreme dose. For angiogenic stimulation with VEGF, this dilemma of delivering VEGF protein active is now attacked in parallel by tissue engineering and somatic gene therapy approaches. These include first, novel, "smarter" biomaterials that present the polypeptide growth factors properly and
Lifeline Research Meeting Abstracts 1297
Volume31, Number6
against expressed tissue factor. In vitro and in vivo testing of this molecule suggests it retains the properties of both parents and has enhanced effect on retention of the glycoprotein I I b / I I I a receptor blocker compared with the receptor blocker alone. Methods Murine antirabbit G P I I b / I I I a receptor antibody (AZ-1) was conjugated with the antirabbit tissue factor antibody (AP-1) using chemical recombination of antibody fragments. The activity of this novel bispecific product was tested using platelet aggregometry and in a clotting assay designed to measure the inhibitory effects on platelet GPIIb/IIIa receptors and tissue factor respectively. In an ex vivo model the conjugate was delivered to rabbit aorta segments using a microporous infusion catheter (Cordis Corporation, US). The conjugate was mounted in a perfusion circuit to measure retention over 12 hours compared with the original anti-GPIIb/IIIa antibody. Results The conjugate inhibited adenosine diphosphate (ADP) induced platelet aggregation almost completely at a concentration of 45 btg per 108 platelets. It also prolonged tissue factor induced clotting time fourfold at a concentration of 6.5 btg/mL. The retention o f the conjugate was enhanced; 21.8% remained in the vessel segments after 12 hours compared with 10.6% in the control group (P < .05). Conclusion It is possible to construct a conjugate of antiG P I I b / I I I a and tissue factor antibodies that retain biological function and have together enhanced retention in the vessel wall. Ongoing in vivo studies and future trials will be presented. Our goal is to identify for any specific process a new antigen or exposure of previously hidden vessel wall component that will allow us to construct molecules that can be delivered systemically in less than normal systemic doses such that they target the pathology, the so called "magic bullet." Dealing with the problems of thrombolysis resistance, thrombolytic failure, and reocclusion are the current aims of our studies. DEVELOPMENTS
IN CATHETER
TECH-NOLOGIES
Steven Bailey, MD, FACC University of Texas Health Sciences Center at San Antonio San Antonio, Tex
Percutaneous coronary interventions, despite the use of endovascular stents, still have limitations due
to early thrombotic events as well as late neointimal proliferation. 1-4 Thrombus formation has been aggressively addressed using new oral and intravenous systemic antiplatelet therapy and improved flow with stcnts. S-7 Despite these newer therapies there are still major problems such as acute myocardial ischemia and saphenous vein graft degeneration. 8-1° Progressive disease in saphenous vein grafts is often with a friable and thrombotic base that continues to represent a challenging scenario for percutaneous interventions. Local delivery of pharmacologic agents is now possible using catheters specifically designed for regional or site specific delivery)l, ~2 These catheters have been developed to allow an increased volume of agent to be infused at a chosen site in an arterial segment. With the use of the properties of passive diffusion or active infusion, these catheters will deliver pharmaceuticals or genes into the arterial wall. Clinical approaches have used the ability of the new catheters to approach this problem in one of three techniques: (1) low-pressure diffusion, (2) low volume infusion, or (3) direct injection into the artery wall or pericardium. While each catheter system has its own unique characteristics, the current delivery systems have equivalent efficiencies of approximately 0.1% to 1.0%.13 While low, this still represents nearly a 100fold increase over that seen with systemic administration or infusion through a guide catheter. The limiting problem has not been identifying a delivery catheter but determining what pharmaceutical agent should be administered and what specific clinical problem we are treating. Coronary dissection and thrombus formation resulting in acute and subacute vessel closure remain significant problems. 14 The decision of which agents to use for local delivery centered on those agents with prior FDA approval. The widespread availability of thrombolytic and antiplatelet therapies resulted in the early use of these agents for local administration. The Dispatch catheter was one of the first systems evaluated in the clinical arena as a local administration device for the indication of coronary thrombus. A nondilating catheter that allows antegrade blood flow while the drug infusion was occurring it also offered the oppornmity for prolonged administration of agents without causing myocardial ischemia. This catheter was extensively characterized in humans by Camenzind et all5 in a novel and elegant manner. Using a continuous infusion of tracer that did not have myocardial uptake with rapid clearance (MAG-3), these authors were able to demonstrate that the actual site of administration
JOURNAL OF VASCULAR SURGERY
1298 Lifeline Research Meeting Abstracts
could be imaged and that the washout from the site could be calculated and standard models applied. The Dispatch catheter has been evaluated in a porcine model for the amount o f intramural deposition o f uroldnase when delivered by different techniques. There are small but significant differences in the delivery efficiency between the different methods o f administration. The clinical use o f these catheters has been reported in over 500 patients to date from registry studies. The earliest studies were those using the first generation catheters. The Infusasleeve registry trials and the local PAMI registries b o t h evaluated the use o f the Infusasleeve catheter in delivering heparin locally. The Local PAMI study evaluated heparin administered as an adjunct to direct angioplasty in the setting o f acute myocardial infarction. This study will be discussed in more detail below under the discussion o f using locally delivered heparin. Subsequent to this laboratory evaluation, the Dispatch catheter has been evaluated in two clinical trials. The first trial was a single center investigation o f uroldnase administration in 13 patients with vain graft disease by Glazier et al.17 Despite the use o f urokinase, two patients developed no fellow in patients with a large thrombus burden. The late clinical follow up was disappointing, with four o f the 12 patients returning with recurrence of clinical symptoms and a fifth patient experiencing late sudden death. The concept o f giving local administration o f urokinase was explored in more detail in the Dispatch and Uroldnase in the Elective treatment o f ghrombus [ D U E T ] trial. This randomized multicenter trial also evaluated the use o f the Dispatch catheter in saphenous vein grafts. It was initiated at the same time that intravenous I I b / I I I a antiplatelet therapies were being investigated. This resulted in a limited enrollment o f 53 patients. The trial was discontinued early because o f an inability to enroll patients as well as an interim analysis that demonstrated a trend toward more complications during the course o f the investigation. Patients who had the dispatch catheter used were more prone to early distal embolization and vascular complications. This study did not address the issue o f I I b / I I I a agent infusion through a local infusion catheter. We have subsequently examined the use ofabciximab using local delivery catheters 18 in a group o f patients with angiographically and angioscopically defined thrombus. The abciximab was delivered as 10 mg in 5 cc over 5 minutes. Angioscopy was performed within 5 minutes after abciximab infusion to
ju~e 2000
confirm the effects o f local abciximab infusion. Precedence for local delivery o f I I b / I I I a agents can be found in the publication by Aggarawal et al. 19 These investigators used a polymer-coated stent to which a mutine antibody to the I I b / I I I a locus was adsorbed. This system was demonstrated to decrease acute platelet deposition in a dose-dependent manner. While the amount o f the antibody was insufficient to inhibit 80% o f platelets it was sufficient to "passivate" the first layer ofplatelets that would have adhered to the stent surface. Unfortunately, this did not result in a reduction in late neointimal proliferation. It is entirely possible that a similar surface passivation can occur when I I b / I I I a agents are delivered using a local delivery catheter system. The time course o f elution has not been characterized for local delivery o f I I b / I I I a agents. Further support for the use o f I I b / I I I a agents is found in the study by Robinson et al.20 They used an external flow loop model to evaluate the usc o f I I b / I I I a agents bound to stents. T h e y demonstrated that there was a marked suppression o f platelet adhesion on stents where the I I b / I I I a agents were applied. Prior studies have also shown that other coatings such as heparin and phosphytidylcholine have decreased both acute platelet deposition and late intimal proliferation.21-24 The administration o f uroldnase has also been investigated in a single-center investigation by Glazier et al. 2s This study evaluated the use o f a hydrogel-coated balloon onto which uroldnase had been adsorbed. The investigators immersed the balloon into a solution o f uroldnase containing 50,000 U / c c . Ninety-five patients were enrolled because o f the presence o f thrombus angiographically (acute MI = 50; postinfarction angina, n = 23; and unstable angina n = 22). While no deaths or emergency bypass procedures occurred, there was a 7.4% incidence o f acute complications with no reflow in three patients, distal embolization in one patient, and late vessel closure in one patient. The late clinical event rate compose d o f death, myocardial infarction, and recurrent angina was 30.5%. This study may represent the fact that thrombolytic therapy increases platelet activation in the setting o f acute ischemia and therefore may increase the risk o f embolization and late platelet driven thrombotic events. This same group 26 used the Dispatch catheter in a small pilot series o f 13 patients with saphenous vein grafts whose average age was 11 years. This population all had thrombosis or occlusion o f aged saphenous, and 12 o f the 13 patients presented with unstable angina pectoris. These patients underwent
JOURNAL OF VASCULAR SURGERY
Volume31, Number 6
infusion o f uroldnase t h r o u g h a dispatch catheter followed by definitive therapy with a balloon inflation or stent placement. The dispatch catheter is a nondilating catheter, and the residual stenosis after uroldnase infusion was still 50% _+ 16%. After definitive therapy, the minimal lesion stenosis fell to 26% _+ 15%. Procedural success was 14 of 15 lesions (12 patients). Two patients had transient no reflow with one patient evolving a n o n - q wave myocardial infarction. Late clinical events were present in four patients with restenosis. A fifth patient with severe three vessel coronary disease had sudden death at home not felt to be related to the therapy. Local delivery ofunfractionated and fractionated heparin has been evaluated. This small molecule has shown promise in prior animal studies in reducing restenosis. Prior animal models including rat, rabbit, and pig 27-3° have suggested that high dose systemic heparin could reduce late neointimal hyperplasia. The use o f low molccular weight heparin was evaluated in the animal model by H o n g et al.31 This study used the channeled balloon to deliver enoxaparin (10 m g / k g ) in a rabbit iliac model. This study demonstrated that acute administration o f a single dose o f low molecular weight heparin was no more effective than control. The combination o f acute administration followed by systemic administration was effective in decreasing late neointimal proliferation. Efficacy o f delivery was only 0.1% to 0.2%. Measurement o f systemic anti-Xa levels demonstrated that the majority o f the enoxaparin was distributed systemically. It is unclear whether the low rate o f local delivery may explain the lack o f efficacy or it is simply not an effective agent for neointimal hyperplasia. It is also unclear if the additional injury from the balloon delivery played a role in the late outcomes. Finally, neointimal hyperplasia has been demonstrated to be a predominate mechanism o f restenosis after stcnt placcment as contrasted with vessel remodeling that occurs after balloon angioplasty, a2 Systemic therapy with nadroparin, a low molecular weight heparin, was attempted in 354 patients in the Fraxiparine Angioplastie Coronaire Transluminale (FACT) trialY There was no difference in either clinical or angiographic outcomes at the 6-month end point. This hypothesis has been tested by Kiesz et at 34 in the Polish Local delivery in the Nir stent trial ( P O L O N I A trial). This trial used low molecular weight heparin (enoxaparin) as a single dose infused into the artery using the transport balloon prior to stent placement. Both the acute outcomes as well as
Lifeline Research Meeting Abstracts 1299
the late clinical and angiographic outcomes were assessed. This trial demonstrated that use o f locally delivered enoxapafin was sufficient to avoid the need for systemic heparinization during the procedure. These data have been presented in preliminary form and demonstrate a significant decrease in neoinfimal hyperplasia as well as the late loss index. These data are the first study to demonstrate angiographic and clinical success using locally delivery in clinical cardiology. There was a publication by Oberhoff et al a5 on a second small series o f patients, in which the results o f local delivery o f intracoronary reviparin in 18 patients were reviewed. The primary success was 100% for delivery o f 7000 units o f intracoronary bolus followed by 1500 units intravenously. These patients also received subcutaneous reviparin 7000 units daily for the next 28 days. The angiographic restenosis rate from this registry was 28% ( 5 / 1 8 ) at 6 months in vessels that were 2.5 to 3.0 mm in diameter. This demonstrates the safety o f this technique in small vessels. It remains to be proved if this technique will reduce restenosis in a randomized trial. Unfractionated heparin has also been evaluated in animal models as a method o f decreasing neointimal hyperplasia. This technique has been tested in several animal models as well as with several delivery catheters. It is unclear what amount ofheparin must be available from local delivery. The only available study with quantitative information is from Mitchel et al. 36 This study demonstrated that the delivery o f only 0.5 to 0.6 units was sufficient to decrease early platelet deposition, but it took 30 to 60 units to attenuate smooth muscle cell proliferation. Clinical studies using heparin therapy delivered using catheter-based systems have been promising. LopezSandon et al37 demonstrated decreased neointimal proliferation at 28 days in a porcine model after infusion using a double balloon catheter. Heparin has been demonstrated to diffuse across the entire arterial wall and into the adventitia. The rate o f platelet deposition varies with the vessel injured. There is a threefold difference in platelet deposition between peripheral vessels and coronary vessels after balloon injury. The hypothesis that limiting acute platelet deposition will decrease late restenosis was the basis for the HIPS trial. This trial randomized 2 5 0 patients between 5000 units o f intracoronary heparin and intraluminal hepafin. The end points were angiographic stenosis as well as intravascular determined plaque volume. This study demonstrated that it was safe to administer heparin in this manner prior to stent implantation. There was no difference in late angiographic or clinical
JOURNAL OF VASCULAR SURGERY
1300 Lifeline Research Meeting Abstracts
outcome in this study. There were no differences in the demographics of the enrolled paticnts between the two groups of intramural versus intraluminal. The ages were the same (57.8 _+ 11.3 vs 59.6 _+10.5 years), male gender was similar (80.2% vs 75.0%) as was the incidence of unstable angina (75.8% vs 73.9%) and prior myocardial infarction (47.3% vs 39.1%). In addition the LVEFs were similar at 54.1% _+9.I% vs 54.3% _+8.1%. This procedure was very safe with a similar percentage o f procedural success o f 99.4% versus 100%. There were no differences in the late clinical outcomes with respect to revascularization. One patient suffered a q wave myocardial infarction and subsequently died in the intramural group, and the incidence o f non-q myocardial infarctions was no different between the groups (8.8% vs 12.5%). Note that the 6-month revascularization rate is exceptionally low in both groups. This is consistent with the large vessels that were entered into this study. The angiographic restcnosis rate was 12.7% in both groups, and there was no difference in late clinical outcomes. A primary end point of the study was intravascular ultrasound determined plaque volume inside the stented segment o f vessel. The intravascular ultrasound-determined vessel volume was 36.4 _+ 20.1 m L 3 in the intramural-treated group and 44.1 _+ 25.1 m L 3 in the intraluminal-control group. The limitations o f the HIPS trial included the use o f a low dose o f heparin for local administration. It is possible that the study dosage was inadequate to achieve sufficient local therapeutic levels. Animal trials had used equivalent doses 10 times higher than those used in this study. Alternatively, this form o f therapy may not be effective in preventing late intimal hyperplasia. This finding was predicted by the studies of Edelman et al38 who demonstrated that diffusion was not effective in treating the thick arteries found in diseased coronary vessels. They showed that adventitial delivery ofheparin was required for suppression ofneointimal proliferation. The use o f unfractionated heparin covalently bound to stents was evaluated in the heparin-coated stent study "Benestent II. "39 This trial evaluated a novel, covalently bound heparin-coated stent compared with balloon angioplasty. The subacute rate was exceptionally low; however, the rate o f restenosis in the stent group was not different than that predicted from the n o n c o a t e d stent used in the B E N E S T E N T I trial. This suggests that there is no antiproliferative effect o f the heparin coating. This mirrors data from animal studies that also failed to demonstrate a decrease in neointimal proliferation in
June 2000
a porcine model. 40 It is unclear whether higher doses o f heparin or noncovalently b o u n d heparin would have decreased the incidence o f neointimal proliferation. Direct thrombin inhibitors have also been used in an attempt to decrease late neointimal proliferation. The trial by Burchenal et al41 o f systemically delivered Bivalirudin (Hirulog) demonstrated that this agent was not effective in decreasing late neointimal proliferation compared with heparin therapy alone in a balloon angioplasty population. The therapy had only a 24-hour duration and is a significant limitation in study design as several studies have shown that thrombin generation c o m m o n l y occurs up to 48 hours after the procedure. This is similar to the data from I I b / I I I a agent's (Capture) trial42 that did not demonstrate a clinical benefit beyond 30 days when only a bolus and one hour infusion ofabciximab were administered after percutaneous interventions. Local delivery o f I I b / I I I a agents is being considered. The exciting data that clinical events are decreased up to 3 years after systemic administration are encouraging. T h e r e are anecdotal reports o f angiographic controlled studies that have demonstrated that single bolus administration o f abciximab decreases restenosis following percutaneous transluminal angioplasty in high risk patients. While this finding is contrary to the data that platelet deactivation requires suppression o f over 80% o f platelets, it may support the fact that restenosis may be more closely tied to the a 3 / b v receptor. Alternatively, "local platelet suppression" at the lesion site may be sufficient to allow adequate platelet inactivation to occur to limit local platelet deposits after an initial layer o f platelets are deposited. This basal layer o f platelets is important in protecting the vessel surface from further events. There are data that support the use o f locally delivered I I b / I I I a agents using stents. The work ofAggarwal et al43 that demonstrated that platelet deposition could be significantly inhibited by a murine antibody to the I I b / I I I a locus is an important finding. The authors found that using polymer-coated stents onto which the AZ1 antibody had been adsorbed inhibited platelet deposition both in vitro and in vivo. The elution curves demonstrated a biexponential curve with an early rapid phase followed by a slow elution such that 40% o f the AZ1 antibody was still present at 14 days. Platelet deposition as assessed by actual platelet counts and physiologic parameters such as cyclic flow variation were improved in the coated stent. Unfortunately, this study did not demonstrate any difference in the late intimal thickness or the
JOURNAL OF VASCULARSURGERY Volume 31, Number 6
intima to media thickness ratio in the A Z l - c o a t e d stent compared with the n o n - A Z T - c o a t e d stent. This finding may represent the fact that the stent had an inadequate a m o u n t o f the I I b / I I I a agent present or more likely the fact that a different receptor blockade may be required. The ability to use locally administered I I b / I I I a agents as single dose therapy for rapid dissolution o f thrombus has been demonstrated in our laboratory. We used angioscopy to evaluate lesions in patients with unstable angina and postinfarction angina to determine the presence ofthrombus. Abciximab was delivered through local delivery catheters such as the Infusasleeve (Local Med, Palo Alto, Calif) or the Dispatch catheter (Scimed). With the use o f only 10 mg o f abciximab, near complete resolution o f thrombus could be demonstrated within 10 minutes using the angioscope. The patient population was too small to allow any information regarding restenosis although only one o f the 11 patients returned with clinical restenosis. Clinical information regarding the use o f other agents such as Taxol is not yet available. This agent does have significant promise in the treatment o f neointimal proliferation. This agent acts in a different manner than agents that decrease cell proliferation. Taxol and other taxenes interfere with the ability o f microtubules. This interferes with the ability o f smooth muscle cells to change shape and move into the artery wall. This agent has been demonstrated by Sollott et al44 to decrease neointimal hyperplasia in a rat model at 11 days in a dose-dependent manner. Subsequent studies have been presented in abstract form that also indicate that this agent can be delivered through stents in a local manner to decrease late neointimal hyperplasia. This therapy has many similarities to the effects o f irradiation and may mimic the cytostatic rather than cytotoxic agents that have previously been used. Unfortunately, restenosis may not be significantly impacted if vessel recoil is not affected by this therapy. Axel et al45 have evaluated the use o f Taxol for local infusion using the microporous infusion catheter from Cordis corporation. This agent had both an in vitro effect as well as a significant effect in a thermal injury rabbit carotid model. The decrease in intimal hyperplasia was significantly greater in the treated animals than in the control animals. The neointimal area was 0.21 + m m 2 vs 0.36 + 0.29 m m 2 in the treated versus the control vessels. Importantly, this decrease was not accompanied by any delay in endothelialization. This differs from the findings seen after irradiation therapy in which the
Lifeline Research Meeting Abstracts 1 3 0 1
endothelial cell was significantly delayed. In addition the concentrations required to inhibit micrombule cross-linking were lO-fold lower than that required to inhibit cell proliferation. These levels are achievable using local delivery techniques. In addition the use o f paclitaxel combined with low dose locally delivery irradiation also appears to have promise as a new therapeutic modality. This technique could potentiate both the microtubular effect as well as the cytotoxic effect o f both therapies. Conclusions Local drug delivery o f thrombolytic agents such as urokinase has been shown to be safe and effective for the treatment o f intracoronary thrombus. The delivery o f GP I I b / I I I a agents is currently being evaluated in the same setting. Early experience suggests that local delivery o f these agents may be as effective as systemic delivery. W h e t h e r long-term benefits will be seen is unclear at this time. Current techniques o f local delivery using conventional doses o f heparin have not been shown to decrease restenosis in a prospective randomized trial or to decrease the amount ofneointimal hyperplasia. There is evidence that new therapies using cell cycle specific agents such as paclitaxel may inhibit cell proliferation and the production o f cxtracellular matrix. Current experience indicates that the choice o f the delivery system is important to the procedural outcomes as seen in the D U E T trial. Careful consideration will need to be given prior to future studies in characterizing the limitations o f delivery system and efficiency prior to embarking on future studies o f local therapy to decrease neointimal proliferation. REFERENCES 1, Hoffmann R, Mintz GS, Dussaillant GR, Popma JJ, Pichard AD, Sailer LF, et al. Patterns and mechanisms of in-stent restenosis: a serial intravascular ultrasound study. Circulation 1996;94:1247-54. 2. Faxon DP, Coats W, Currier J. Remodeling of the coronary artery after vascular injury. Prog Cardiovasc Dis 1997; 40:129-40. 3. Kimura T, Kaburagi S, Tamura T, Yokoi H, Nakagawa Y, Yokoi H, et al. Remodeling of human coronary arteries undergoing coronary angioplasty or atherectomy. Circulation 1997;96:475-83. 4. Kronowski R, Mintz GS, Kent KM, Pichard AD, Sailer LF, Bucher TA, et al. Increased restenosis in diabetes mellitus after coronary interventions is due to exaggerated intimal hyperplasia. Circulation 1997;95:1366-9. 5. Lefkovits J, Ivanhoe RJ, Califf RM, et al. Effects of platelet glycoprotein I I b / I I I a receptor blockade by a chimeric monoclol~al antibody (abciximab) on acute and six-month out-
1302
6.
7.
8. 9.
10.
11. 12. 13. 14.
15.
16.
17.
18.
19. 20. 21.
22.
IOURNAL OF VASCULARSURGERY lune 2000
Lifeline Research Meeting Abstracts
comes after percutaneous transluminal coronary angioplasty for acute myocardial infarction. Epic Investigators Am Cardiol 1996:77:1045-51. Tcheng JE. Glycoprotein I I b / I I I a receptor inhibitors: putting the EPIC, Impact II, Restore and Epilog trials into perspective. Am J Cardiol 1996;78:35-40. Topoi EJ. Prevention of cardiovascular ischennc complications with new platelet glycoprotein IIb/IIIa inhibitors. Am Heart i 1995;130:666-72. Banters C, Isner JM. The biology of restenosis. Prog Cardiovasc Dis 1997;40:107-16. Mintz GS, Popma II, Pichard AD, Kent KM, Satler LF, Wong SC, et al. Arterial remodeling after coronary angioplasty. Circulation 1996;94:35-43. Rogers C, Edelman ER. Endovascular stent design dictates experimental restenosis and flarombosis. Circulation I995; 91:29953001. Bailey SR. Local drug delivery: current applicatioris. Prog Cardiovasc Dis 1997;40:183-204. Brieger D, Topol E. Local drug delivery systems and prevention ofrestenosis. Cardiovasc Res 1997;35:405-13. Edelman ER, Lovich M. Drug delivery models transported to a new level. Nature Biotechnol 1998;16:136-7. Sehieman G, Cohen BM, Kozina l, Erickson JS, Podoloin RA, Peterson KL, et al. Intracoronary urokinase for intracoronary thrombus accumulation complicating percutaneous intraeoronary thrombus accumulation complicating percutaneous transluminal coronary angioplasty in acute in acute ischemic syndromes. Circulation 1990;82:2052-60. Camenz~fid E, Bakker A, Reijs A, Vangeijlswijk IM, Boersma E, Kutryk MJB, et al. Site-specific intracoronary heparin delivery in humans after balloon angioplasty: a radioisotopic assessment of regional pharmacokinetics. Circulation 1997; 96:154-65. Mitchet JF, Shwedick M, Alberghini TA, Iqa-fibbs D, McK~y RG. Catheter-based local thrombolysis with uro!dnase: comparative efficacy of intraluminal clot lysis with conventional urokinase infusion techniques in an in vivo porcine thrombus model. Cathet Cardiovasc Diagn 1997;41:293-302. Glazier JJ, Keman FJ, Bauer HI-I, Gram DB, Primiano, CA, Mitchel JF, et al. Treatment of thrombotic saphenous vein bypass grafts using local uroldnase infusion therapy with the dispatch catheter. Cathet Cardiovasc Diagn 1997;41:261-7. Bailey SR, O'Leary E, Chitton R. Angioscopic evaluation of site-specific administration of ReoPro. Cathet Cardiovasc Diagn 1997;42:181-4. Aggarwal RK, Ireland DC, Azrin MA, Ezekowitz MD, de Bono DP, Gershlick AH. Circulation 1996;94;3311-8. Robinson et al article on IIb/IIIa and flow loop of stents. DeScheerder IK, Wilczek KL, Verbeken EV, Vandorpe J, Lan PN, Schacht E, et al. Biocompatability of polymer-coated oversized metallic stents implanted in normal porcine coronary arteries. Elsevier Science Atherosclerosis 1995;114:105-14. Hubbell JA. Biomatetials in tissue engineering. Biotecfmology
1995;13:565-76. 23. Mckenna CJ, Camrud AR, Sangiorgi G, Kwon HM, Edwards WD, Holmes DR Jr, et al. Fibrin-film stenting in a porcine coronary injury model: efficacy and safety compared with uncoated Stents. J Am Coil Cardiol 1998;31:1434-8. 24. Lincoff AM, Furst JG, Ellis SG, Turch RJ, Topoi EJ. Sustained local delivery of dexamethasone by a novel intravascular eluting stent to prevent restenosis in the porcine coronary injury model. J Am Coil Cardiol 1997;29:808-16. 25. Glazier JJ, Hirst JA, Keman FJ, Fram DB, Eldin AM,
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38. 39.
40.
Primiano CA, et al. Site-specific intracoronary thrombolysis with urokinase-coated hydrogel balloons: acute and followup studies in 95 patients. Cathet Cardiovasc Diagn 1997; 41:246-53. Glazier Jl, Kernan FJ, Bauer FIH, Fram DB, Primiano CA, Mitchel JF, et al. Treatment of thrombotic saphenous Vein bypass grafts using local urokinase infusion therapy with the dispatch catheter. Cathet Cardiovasc Diagn 1997;41:261-7. Guyton JR, Rosneber RD, Clowes AW, Karnowsky MJ. Inhibition of rat arterial smooth muscle cell proliferation by hepatin. Circ Res 1980;46:625-34. Clowes AW, Clowes MM. Kinetics of cellular proliferation after arterial injury, II: inhibition of smooth muscle growth by heparin. Lab Invest 1985;52:611-16. Wolinsky H, Thung SN. Use of a perforated catheter to deliver concentrated heparin into the wall of the normal into the wall of the normal coronary artery. J Am Coll Cardiol 1990;15:475-81. Azrin MA, Mitchel IF, Fram DB, Pedersen CA, Cartun RW, Barry JJ, et al. Decreased platelet deposition and smooth muscle cell proliferation following intramural heparin delivery with Hydrogel-coated balloons. Circulation 1994;90: 433-41. Hong MK, Wong SC, Barry Jl, Bramwell O, Tjurmin A, Leon MB. Feasibility and efficacy of locally delivered enoxaparin via the channeled balloon catheter on smooth muscle cell proliferation following balloon injury in rabbits. Cathet Cardiovasc Diagn 1997;41:232-40. Hoffmann R, Mintz GS, Dussaillant Gtl, Popma JJ, Pichard AD, Satler LF, et al. Patterns and mechanisms of in-stent restenosis: a serial intravascular ultrasound study. Circulation I996;94:1247-54. Lablance JM, McFadden EP, Meneveau N, Lusson JR, Bertrand B, Metzger IP, et al. Effect of dadroparin, a lowmolecular weight heparin on clinical and angiographic restenosis after coronary balloon angioplasty: The FACT Study. Circulation 1997;96:3396-402. Kiesz RS, Buszman P, Martin JL, Deutsch E, Rozek NIM, Rweieki M, et al. Polish-American Local Lovenox Nir Stent Assessment Study (POLONIA): final results [abstract]. J Am Coll Cardiol 1999;33 Suppl A:4A. Oberhoff M, Banmbach A, Hermann T, Diehl C, Maier R, Athanasiadis A, et al. Local and systemic delivery of low molecular weight heparin following PTCA: acute results and 6-month follow-up of the initial experience with the porous balloon (PILOT-study). Preliminary Investigation of Local Therapy Using Porous PTCA Balloons. Cathet Cardiovasc Diagn 1998;44:267-74. Mitchel JF, Azrin MA, Fram DB, Lanrine MB, McKay RG. Localized delivery of heparin to angioplasty sites with iontophoresis. Cathet Cardiovasc Diagn 1997;41:315-23. Lopez-Sandon J, Sobrino N, Bamallo C, Lorenzo A, Jimenez l, Calvo L, et al. Locally delivered heparin reduces intimal hyperplasia and lumen stenosis following balloon injury in swine [abstract]. Eur Hear J Suppl 1993;14:191. Edelman ER, Lovich M. Drug delivery models transported to a new level. Nat Biotechnol 1998;16:136-7. Serruys P, Van Hour B, Bonnier H, Legrand V, Garcia V, Macaya C, et al. Randomized comparison of implantation of heparin-coated stents with balloon angioplasty in selected patients with coronary artery disease (Benestent II). Lancet 1998;352:673-81. Hardhammar PA, van Beusekom I-IM, Emanuelsson HU, Hoffman SH, Albertson PA, Verdouw PD, et al. Reduction
JOURNAL OF VASCULARSURGERY
Volume31, Number6
41.
42.
43.
44.
45.
in thrombotic events with heparin-coated Palmaz-Schatz stents in normal porcine coronary arteries. Circulation 1996;93:423-30. Burchenal JEB, Marks DS, Mann JT, Schweiger MJ, Rothman MT, Ganz P, et al. Effect of direct thrombin inhibition with Bivalirudin (Hirulog) on restenosis after coronary angioplasty. Am J Cardiol 1998;82:511-5. van de WerfF. More evidence for a beneficial effect ofplatelet glycoprotein IIb/IIIa-blockade during coronary interventions: latest results from the EPILOG and CAPTURE trials. Circulation 1996;93:637. Aggarwal RK, Ireland DC, Azrin MA, Ezekowitz MD, deBono DP, Gershlick Ah. Antithrombotic potential of polymer-coated stents eluting platelet glycoprotein I I b / I I I a receptor antibody. Circulation 1996;94:3311-7. Sollott SJ, Cheng L, Pauly RR, Jenkins GM, Monticone RE, Kuzuya M, et aft. Taxol inhibits neointimal smooth muscle cell accumulation after angioplasty in the rat. J Clin Invest 1995;95:1869-76. Axel DI, Kunert W, Goggelmann C, OberhoffM, Herdeg C, Kuttner A, et al. Paclitaxel inhibits arterial smooth muscle cell proliferation and migration in vitro and in vivo using local drug delivery. Circulation 1997;96:636-45.
FIBRIN-BASED MATRICES F O R A N G I O GENIC STIMULATION Andreas H. Zisch, MD U. Schenk B. Schwe#zer S. Sakiyama-Elbert J. C. Schense jr. A. Hubbell Inst#ute for Biomedical Engineering Zurich, Switzerland
Our laboratory's research in vascular tissue engineering focuses on two distinct goals: first, in the context of building blood vessels dc novo, we aim to engineer biomaterial surfaces that are able to attract and retain endothelial cells organized as two-dimensional endothelial cell sheet under flow conditions, similar to the the native luminal vessel wall. Second, featuring the three-dimensional organization of endothelial cells, the generation of biomaterial matrices, which when grafted to the wounded or ischemic tissue, locally stimulate neovessel formation in the surrounding of the graft and also in the graft itself. In both cases, the material designs aim the vascular grafts to become populated in situ with endothelial cells derived from the body's endogenous sources (ie, from the artery adjacent to the graft, from existing blood vessels, or from circulating endothelial progenitor cells). For that reason, these grafts must have properties specifically attractive for endothelial cells. Through basic research, a better knowledge has been developed about the molecular aspects and requirements for reendothe-
Lifeline Research Meeting Abstracts 1303
lialization and blood vessel formation. This biological information can be inferred on a biomaterial by incorporating peptide or protein motifs that are necessary for the recruitment and the binding of endothelial cell binding. Polypeptide growth factors, in particular vascular endothelial growth factor (VEGF), have evolved as critical for many aspects of endothelial cell biology, such as the maturation, the proliferation, the assembly, and the survival of endothelial cells. The Hubbell laboratory has recently developed methodology that allows fibrin to be modified by covalent incorporation of bioactive peptides or proteins during the coagulation process. The base matrix fibrin is a natural ingrowth matrix for endothelial cells. To confer the angiogenic features of VEGF upon fibrin, we used this methodology to covalently incorporate VEGF121, an otherwise diffusible VEGF isoform into the fibrin matrix. This VEGF121/fibrin matrix exhibits strong angiogenie responses both in vitro and in vivo as described below. Further, we will discuss a sophisticated and elegant fibrin modification scheme in which growth factors, such as the VEGF165 isoform, can be entrapped and retained within the matrix through their heparin-binding abilities. From these depots, growth factors can be released by controllable diffusion as well as the activity of matrix-degrading enzymes secreted by cells invading the matrix. The clinical demand for angiogenic materials-modified fibrin as material for clinics Design of materials that induce angiogenesis in vivo is important for many areas of tissue regeneration and wound healing. One of the more relevant and therefore well studied needs is in treatment of chronic wounds, such as diabetic ulcers, pressure ulcers, or venous static ulcers that are caused by deficient vascular supply. Scores of trials with ulcer patients have been conducted over the past years using growth factor proreins or cytokines as therapeutics, with many discouraging results. One likely explanation for the failure of many therapies is the current modes of growth factor delivery in the context of creams and sprays, which are often inappropriate for providing sustained delivery of active factors. Without enbedment, however, the soluble protein factors are cleared very rapidly from the target site, or never reach the target in sufficient quanrifles, or reach it only at extreme dose. For angiogenic stimulation with VEGF, this dilemma of delivering VEGF protein active is now attacked in parallel by tissue engineering and somatic gene therapy approaches. These include first, novel, "smarter" biomaterials that present the polypeptide growth factors properly and