Novel b-emitting poly(ethylene terephthalate)surface modification

Xin Qu, Judah WeinbergerInterventional Cardiology Center, Departments of Medicine and Pharmacology, Columbia University, 161 FortWashington Avenue, AP-551, New York, New York 10032

Received 23 March 2000; accepted 12 April 2000

Abstract: Restenosis after percutaneous interventions incoronary and peripheral arteries leads to repeat proceduresand surgery in a significant number of patients. We havepreviously demonstrated that irradiation of an arterial siteusing an endovascular source (brachytherapy) is highly ef-fective in preventing the restenotic process. To this end, anovel beta radiation delivery system was developed, basedon the adsorption of 32P (o-phosphoric acid) by pH-sensitivechitosan hydrogel on a poly(ethylene terephthalate) (PET)balloon surface. The PET balloon surface was treated withoxygen plasma and coated with chitosan hydrogel. Covalentbonds, ionic bonds, and hydrogen bonds all contribute to theadhesion between chitosan hydrogel and PET. In the aque-ous phosphoric acid (PA) solution, the –NH2 groups of chi-tosan were protonated by PA and the adsorption of PA oc-

curred at the same time. The effect of PA concentration andtemperature on adsorption efficiency and kinetics werestudied. More than 70% PA was adsorbed on the samplesurface in 0.2 mM PA solution. The surface of samples wasalso investigated by attenuated total reflection–Fouriertransform infrared spectroscopy and scanning electron mi-croscopy. PET surface may be modified to carry high activitybeta emitters; such materials may be useful in a therapeuticsetting. © 2000 John Wiley & Sons, Inc. J Biomed Mater Res,52, 492–497, 2000.

Key words: brachytherapy; intracoronary radiation; PET;balloon; chitosan; o-phosphoric acid; D,L-lactic acid; surfacemodification

INTRODUCTION

Balloon angioplasty and stent implantation arewidely applied therapies for symptomatic, obstructiveatherosclerotic coronary artery, and peripheral arterialdisease. A major source of morbidity, and a majorremaining limitation, restenosis rates are as high as 30to 50% with these procedures.1,2 Pharmacological andmechanical approaches to restenosis prevention thusfar have been disappointing.3,4 The impact on reste-nosis achieved with intracoronary stents is bluntedbecause of a persistent neointimal proliferative re-sponse. We have demonstrated previously the abilityof intravascular sources of ionizing radiation to pre-vent neointimal proliferation in models of restenosis.Intravascular radiation therapy, or brachytherapy,seems to prevent restenosis by reduction of smoothmuscle cell proliferation, matrix formation, and byminimizing the late constriction of the vessel wall.The effects of ionizing radiation on cell proliferation

and vascular remodeling were demonstrated previ-ously in several animal studies5–9 and in early clinicaltrials.10–12

A number of platforms have been devised to deliverbrachytherapy from catheter-based systems (high-dose rate) or radioactive stents (low-dose rate). Beta-emitters delivered by catheter-based approaches in-clude 90Y wire sources (Schneider), encapsulated90Sr/Y (Novoste), and 32P seeds (Guidant), and 188Reas a solution source for balloon inflation.13 192Iridium,a gamma emitter, is being developed in wire-affixedseed geometry (Cordis J & J). For reasons of shieldingand patient and operator safety, a clear preference ex-ists for beta sources, although the relative efficacy ofvarious isotopic sources is still under investigation.

Chitosan, (1,4)-2-amino-2-deoxy-b-D-glucan, is anatural polymer generally obtained by extensivedeacetylation of chitin isolated from crustacean shells.Because of its special biological, chemical, and physi-cal properties, chitosan and its derivatives have appli-cations in many industrial, agricultural, and biomedi-cal activities.14,15 Chitosan hydrogels synthesizedfrom chitosan and D,L-lactic acid have been reportedin the literature.16,17 The free amino groups and po-rous structure of chitosan hydrogels provide them the

Correspondence to: J. Weinberger; e-mail: jzwl@columbia.edu

© 2000 John Wiley & Sons, Inc.

ability to adsorb o-phosphoric acid (PA) in aqueoussolution.

In this article, we report a novel beta radiation de-livery material that may be useful in a particularlysimple brachytherapy delivery system, based on theadsorption of 32P (o-phosphoric acid) by a chitosanhydrogel on poly(ethylene terephthalate) (PET) sur-face. PET surface is found in existing angioplasty bal-loons, and could also be used in wire geometry. Weinvestigate the variables affecting both efficiency ofisotope uptake and surface capacity, as well as themorphology of the surface.

MATERIALS AND METHODS

Chitosan (Mw = 150 kD) from Fluka (Buchs, Switzerland)and D,L-lactic acid (85%) from Fisher Scientific (Pittsburgh,PA, USA) were used for preparation of chitosan hydrogel.O-phosphoric acid (85%) (Fisher Scientific, Pittsburgh, PA)was used for adsorption experiment. 32P-o-phosphoric acid(8500–9120 Ci/mmol, 10 mCi/mL) (NEN, Boston, MA) wasused for measuring adsorption efficiency and kinetics byliquid scintigraphy. Ecoscint A was obtained from NationalDiagnostics, Inc. (Atlanta, GA, USA). PET balloons (15 × 55mm) were obtained from Advanced Polymers, Inc. (Salem,NH, USA).

Surface modification of PET balloons bychitosan hydrogel

PET balloons were treated by oxygen plasma to obtainhydrophilic surfaces. The PET balloons were then cut intosmall pieces (5 × 10 mm, thickness 30 ± 2 mm). These PETfilms were coated on one side surface by 1% chitosan/D,L-lactic acid solution, which was prepared by dissolving chi-tosan powder in D,L-lactic acid aqueous solution with theweight ratio of chitosan/lactic acid = 1⁄2. The coated filmswere dried in an oven at 80°C for 1 h. The thickness of thecoated films was 31 ± 2 mm. The synthesis and characteriza-tion of these chitosan hydrogels have been described previ-ously.16,17

Adsorption of PA by hydrogel layer

Coated films were immersed in varying concentration ofthe trace 32P radiolabeled o-phosphoric acid aqueous solu-tions (0.5 mL) at room temperature or at 50°C. The adsorp-tion efficiency and kinetics were computed by measuringthe residual 32P in the adsorbing solutions. Adsorption effi-ciency of samples at time t was calculated by the followingequation:

Adsorption efficiency (%) = (M0 − Mt)/M0 (×100)

where M0 and Mt are the amount of PA in solutions beforethe adsorption (M0) and after the adsorption at time t (Mt).

Characterization

Fourier transform infrared (FTIR) transmission spectrawere obtained using the attenuated total reflection (ATR)technique. The surfaces were analyzed on a Perkin-Elmer2000 infrared Fourier transform spectrometer. Scanning elec-tron microscopy (SEM) analyses were performed using aJeol JSM-5600LV scanning electron microscope (Japan).Samples were mounted on metal stubs and sputter-coatedwith gold-palladium. A liquid scintillation counter fromLKB Wallace, 1209 Rackbeta, was used to measure the con-centration of the 32P isotope in the solution before and afterthe adsorption. A 10-mL solution was taken and mixed witha 5-mL scintillation solution. The thickness of the films wasmeasured with a micrometer caliper.

RESULTS

Chitosan hydrogel layer on PET balloon

Untreated PET balloon surface is hydrophobic. Af-ter oxygen plasma treatment, the surface becomes hy-drophilic. Figure 1(A) shows the morphology of theplasma-treated PET film. The parallel lines on the sur-face reflect the manufacturing process. After coatingwith chitosan hydrogel, the PET film became muchsmoother and a thin layer, about 1mm, appeared onthe surface [Fig. 1(B)]. Figure 2 shows the ATR–FTIRspectra of the untreated and oxygen plasma-treatedPET film. Compared with the spectrum of the un-treated sample in Figure 2(A), the oxygen plasma-treated sample [Fig. 2(B)] has two new small peaksappearing at 3610 cm−1 and 3530 cm−1 which are at-tributed to the –COOH and –OH groups formed onthe surface by oxygen plasma treatment. As shown inFigure 3(A), these two peaks were overlapped by abroad peak at 3250 cm−1 assigned to the chitosan hy-droxyl groups, after the chitosan hydrogel was coatedon the surface. In addition, a new peak correspondingto free amino groups of chitosan appeared at 1563cm−1 and the peak at 1067 cm−1 was assigned to thechitosan saccharide structure.

Adsorption of PA

After the chitosan hydrogel coated film was im-mersed in 2 mM of PA solution for 2 h and dried; theATR–FTIR spectrum of the sample surface was mea-sured. Two new and strong peaks (1609 cm−1 and 1530cm−1), that are related to the deformation of NH3

+

groups in chitosan, appeared in the spectrum [Fig.3(B)]. The –PO4 vibration peaks are predominant com-pared with the existing chitosan peaks. The absorption

493b-EMITTING PET SURFACE MODIFICATION

peaks at 1130 cm−1 and 1020 cm−1 are due to the P–Ostretch. Meanwhile, the peak of hydroxyl group at3250 cm−1 increased after forming the chitosan phos-phate. Figure 4 shows the SEM photograph of the filmafter the adsorption. The film is a little rough as com-pared with the CS-coated film and with some smallwhite particles on the surface. The upper left cornerdisplays disruption of the chitosan coating caused byhandling. This allows an estimate of the film thicknessto be approximately 1 mm.

Adsorption efficiency of chitosan hydrogel layer

The PA adsorption efficiency of the treated PET sur-face was examined at room temperature and at 50°Cas a function of PA concentrations. All data were mea-sured after the samples were immersed in the solu-tions for 2 h. As shown in Figure 5, the adsorptionefficiency of PA increases with the decrease of solutionconcentration and reaches maximum at 0.2 mM. In

general, the efficiency values at 50°C are higher thanthose values at room temperature, whereas two curveshave the same trend. As shown in Figure 6, theamount of H3PO4 adsorbed on the sample surfacesincreases with the increase of the solution concentra-tion. Both curves level off at 5 mM because of thesaturation of H3PO4 adsorbed on sample surface. Thefilm adsorbed more H3PO4 at higher temperature.

Figure 7 presents the adsorption kinetics of samplesin different PA concentrations at room temperature. Itis clear that the adsorption efficiency notably dependson the solution concentration and the adsorption time.

Figure 1. SEM photographs of (A) oxygen plasma-treatedPET surface and (B) chitosan hydrogel-coated PET surface.

Figure 2. ATR–FTIR spectra of (A) untreated PET surfaceand (B) oxygen plasma-treated PET surface.

Figure 3. ATR–FTIR spectra of (A) chitosan hydrogel-coated PET surface, (B) the coated PET surface after adsorp-tion in 2-mM PA solution for 2 h.

494 QU AND WEINBERGER

The higher the concentration of PA solutions, thelower the adsorption efficiency, but the time to equi-librium is much shorter. In all cases, the equilibriumadsorption was almost reached within 2 h.

DISCUSSION

Chitosan hydrogel could not be attached to the un-treated PET balloon surface because untreated PET ishydrophobic. After the oxygen plasma treatment,functional groups such as –OH and –COOH were cre-ated on the sample surface. As illustrated in Figure 8,a chitosan hydrogel coating could be expected becauseof the formation of covalent bonds, ionic bonds, and

hydrogen bonds between the functional groups on thesurface and –NH2 groups of chitosan during the heat-ing process. A similar method has been used to coatpolypropylene films with chitosan for improving dye-ing behavior.18

As shown previously,16,17 aqueous solutions of chi-tosan and lactic acid can form hydrogels after heating.Chitosan is first dissolved in lactic acid solution toform chitosan lactate salt. The dehydration of chitosanlactate salt will occur to form amide groups duringheating. Simultaneously, the polycondensation of lac-tic acid occurs to form lactic acid side chains. The for-mation of chitosan hydrogel is due to the physicalcrosslinking through hydrophobic side chains aggre-gation and intermolecular interactions by hydrogenbonds between side and main chains, which eventu-

Figure 4. SEM photographs of chitosan hydrogel-coatedPET balloon surface after adsorption in PA solution for 2 h.

Figure 5. The effect of temperature and solution concen-tration on sample adsorption efficiency.

Figure 6. The effect of temperature and solution concen-tration on the amount of PA adsorbed on the sample surface.

Figure 7. The adsorption kinetics of samples in the PAsolutions with different concentrations at room temperature.

495b-EMITTING PET SURFACE MODIFICATION

ally lead to a corresponding decrease of chitosan chainmobility in the aqueous solutions.

The chitosan hydrogel layer on the PET surface issmooth and transparent. It is pH sensitive and swellsextensively in aqueous PA solution. The unreactedamino groups of chitosan are ionized by the PA, andthe acid attached to the gels by the ionic bonds. Theinfluence of H3PO4 concentration on the adsorptionefficiency could be divided into three periods. At con-centrations higher than 5-mM solutions, the hydrogellayer adsorption was maximal at 1.8 × 10−3 mmolH3PO4/cm2 at 25°C and 2.5 × 10−3 mmol H3PO4/cm2

at 50°C, which is independent of the concentration.The adsorption efficiency decreases with the increaseof the concentration. In the solution concentration be-tween 0.2 mM and 5 mM, the hydrogel layer adsorbedmore at higher H3PO4 concentration, whereas the ad-sorption efficiency decreased with the increase of con-centration. The efficiency reaches the highest values(70–80%) at 0.2-mM solution. Below this concentra-tion, the adsorption efficiency decreases with the con-centration. As indicated in Figure 7, the adsorptionprocess is always faster in higher H3PO4 concentrationsolutions.

Radiation dosimetry is critical in vascular brachy-therapy. Overdose and underdose could result in re-duction of the treatment effectiveness and may in-crease the radiation toxicity. The radioisotope capacityof this system has been computed as follows. The car-rier-free activity of 32P isotope we used is 8500–9120Ci/mmol, so the maximum achievable activity densityis 20 Ci/cm2 on the PET surface. Typical activitiesneeded for the intracoronary radiation is about 20mCi/cm2 for a 2-cm treatment length. This calculationindicates that surface capacity is approximately 1000-fold greater than required.

The feasibility of providing uniform dosimetry isalso critical in vascular brachytherapy. In usual wire-based delivery procedures, the source is not centeredin available lumen. The lack of centering may be as-sociated with overdosing on one side of the vessel

wall and underdosing on the other side. In this case,the use of a radioactive balloon surface would providethe best attainable uniformity of dose at the arterialwall even if the balloon or vessel takes a turn. Anotherconfiguration of the radioactive surface allowed by thehigh capacity for radioisotope would be as a flexiblepolymer wire coated with the radioisotope.

When the radioactive balloon is inserted into thebody, it is absolutely necessary to make sure the iso-tope will not elute from the balloon into aqueous en-vironment by the blood. The control of 32P-PA off-ratein aqueous solutions by additional coatings is nowunder investigation and will be reported separately.

We have examined the capacity of approach de-scribed herein to other radioisotopes and other poly-mer surfaces. The surface modification of nylon bal-loon after plasma treatment was also achieved by thismethod and similar results have been obtained in ourlaboratory (data not shown). Meanwhile, the chitosanhydrogel on the surface has the ability to chalet radio-isotope ions such as 188Re, which provides anotheralternative to radioactive balloon surface (data notshown).

CONCLUSION

A novel beta radiation delivery system was devel-oped for intracoronary radiation, which is based onthe adsorption of 32P (o-phosphoric acid) by pH-sensitive chitosan hydrogel on PET balloon surface.The radiation dose of the balloons could be manipu-lated by changing the 32P concentration in the solutionand also the proportion of cold and hot PA. The con-trol of 32P-PA off-rate in aqueous solutions from thissystem requires further study.

References

1. Gruentzig AR, King SB, Schlumpf M, Siegenthaler W. Long-term followup after percutaneous transluminal coronary an-gioplasty. N Engl J Med 1987;316:1127–1132.

2. Nobuyoshi M, Kimura T, Nosaka H, Mioka S, Veno K, YokoiH, Hamasaki N, Horiuchi H, Ohishi H. Restenosis after suc-cessful percutaneous transluminal coronary angioplasty: Serialangiographic follow-up of 229 patients. J Am Coll Cardiol1988;12:616–623.

3. Muller DWM, Ellis SG, Topol EJ. Colchicine and antineoplastictherapy for prevention of restenosis after percutaneous coro-nary interventions. J Am Coll Cardiol 1991;17:26(B)–31(B).

4. Urban P, Buller N, Fox K, Shapiro, Bayliss, Rickards. Lack ofeffect of warfarin on the restenosis rate or on clinical outcomeafter balloon coronary angioplasty. Br Heart J 1988;60:485–488.

5. Wiedermann JG, Marboe C, Amols H, Schwartz A, WeinbergerJ. Intracoronary irradiation markedly reduces restenosis afterballoon angioplasty in a porcine model. J Am Coll Cardiol1994;23(6):1491–1498.

6. Wiedermann JG, Marboe C, Amols H, Schwartz A, WeinbergerJ. Intracoronary irradiation markedly reduces neointimal pro-

Figure 8. Surface modification of PET balloon by chitosanhydrogel.

496 QU AND WEINBERGER

liferation after balloon angioplasty in swine: Persistent benefitat 6-month follow-up. J Am Coll Cardiol 1994;25(6):1451–1456.

7. Mazur W, Ali MN, Khan MM, Dabaghi SF, DeFelice CA, Para-dis JP, Butler EBA, Wright E, Fajardo LFB, French A, RaiznerAE. High dose rate intracoronary radiation for inhibition ofneointimal formation in the stented and balloon-injured por-cine models of restenosis: Angiographic, morphometric, andhistopathologic analysis. Int J Radiat Oncol Biol Phys1996;36(4):777–788.

8. Waksman R, Robinson KA, Crocker IR, Gravanis MB, CipollaGD, King SR. Endovascular low-dose irradiation inhibits neo-intima formation after coronary artery balloon injury in swine:A possible role for radiation therapy in restenosis prevention.Circulation 1995;91(5):1533–1539.

9. Verin V, Popowski Y, Urban P, Belenger J, Redard M, Costa M,Widmer MC, Rouzaud M, Nouet P, Grob E, Schwager M,Kurtz JM, Rutishauser W. Intra-arterial beta irradiation pre-vents neointimal hyperplasia in a hypercholesterolemic rabbitrestenosis model. Circulation 1995;92:2284–2290.

10. Condado JA, Waksman R, Gurdiel O, Espinosa R, Gonzalez J,Burger B, Villoria G, Acquatella H, Crocker IR, Seung KB, Lip-rie SF. Long-term angiographic and clinical outcome after per-cutaneous transluminal coronary angioplasty and intracoro-nary radiation therapy in humans. Circulation 1997;96:727–732.

11. Teirstein PS, Massullo V, Jani S, Popma JJ, Mintz GS, Russo RJ,Schatz RA, Guarneri EM, Steuterman S, Morris NB, Leon MB,Tripurareni P. Catheter-based radiotherapy to inhibit stenosisafter coronary stenting. N Engl J Med 1997;336:1697–1703.

12. Verin V, Urban P, Popowski Y, Schwager M, Nouet P, DorsazPA, Chatelain P, Kurtz JM, Rutishauser W. Feasibility of intra-coronary b-irradiation to reduce restenosis after balloon angio-plasty: A clinical pilot study. Circulation 1997;95:1138–1144.

13. Weinberger J. Intracoronary radiation using radioisotope solu-tion-filled balloons. Herz 1998;23:366–372.

14. Muzzarelli RAA. Chitin. Oxford: Pergamon; 1977.

15. Roberts GAF. Chitin chemistry. Houndmills, UK: Macmillan;1992.

16. Qu X, Wirsen A, Albertsson AC. Structural change and swell-ing mechanism of pH-sensitive hydrogels based on chitosanand D,L-lactic acid. J Appl Polym Sci 1999;74(13):3186–3192.

17. Qu X, Wirsen A, Albertsson AC. Synthesis and characteriza-tion of pH-sensitive hydrogels based on chitosan and D,L-lactic acid. J Appl Polym Sci 1999;74(13):3193–3202.

18. Rochery M, Lam TM, Crighton JS. FTIR and ATR analyses ona polypropylene (PP) surface after plasma treatment in thestudy of chitosan surface grafting to improve PP dyeing be-havior. Macromol Symp 1997;119:277–282.

497b-EMITTING PET SURFACE MODIFICATION