Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Aug 17.
Published in final edited form as: J Control Release. 2010 May 8;146(1):23–30. doi: 10.1016/j.jconrel.2010.05.005

Correlation of tissue drug concentrations with in vivo magnetic resonance images of polymer drug depot around arteriovenous graft

Shawn C Owen 1, Huan Li 2, William G Sanders 3, Alfred K Cheung 4, Christi M Terry 5
PMCID: PMC2942017  NIHMSID: NIHMS211229  PMID: 20457189

Abstract

Sustained delivery of anti-proliferative drugs to the perivascular area using an injectable polymeric platform is a strategy to inhibit vascular hyperplasia and stenosis. In this study, the concentrations of sirolimus in vascular tissues were evaluated after delivery using an injectable platform made of poly(lactic-co-glycolic acid)-polyethylene glycol-poly(lactic-co-glycolic acid) (PLGA-PEG-PLGA). In order to optimize the drug release profile, the effect of two solvents or solid loading of the sirolimus into the polymer gel was first examined in vitro. The early release was slower with loading of dry drug into the polymer, compared to drug dissolution in solvents. Dry sirolimus was therefore used to load the polymer and applied to the perivascular surface of the graft-venous anastomosis at the time of surgical placement of a carotid-jugular synthetic hemodialysis graft in a porcine model. This was replenished by ultrasound-guided injection of additional drug-laden polymer at one, two and three weeks post-operatively. Magnetic resonance imaging (MRI) using pulse sequences specifically designed for optimal detection of the polymeric gel showed that the polymer injected post-operatively remained at the juxta-anastomotic perivascular site at two weeks. Sirolimus was extracted from various segments of the juxta-anastomotic tissues and the drug concentrations were determined using HPLC-MS/MS. Tissue sirolimus concentrations at one and two weeks were highest near the venous anastomosis, which were approximately 100- to 500-fold greater than the concentrations necessary to inhibit vascular smooth muscle cell proliferation placement of a carotid-jugular synthetic hemodialysis graft in a porcine model in vitro. Drug concentrations remained above the inhibitory concentrations for at least six weeks postoperatively. Thus, serial injections of sustained-delivery polymer gel loaded with sirolimus can provide high localized concentrations at target vascular tissues and thus may be useful for the prevention and treatment of vascular proliferative disorders such as hemodialysis graft stenosis. In addition, MRI is useful for the monitoring of the location of the drug depot.

1. Introduction

Thermosensitive hydrogels have been investigated for several decades [1–4] as vehicles for controlled drug delivery. In that time, the physical properties of several dozen of these block copolymers have been elucidated, including water content, cross-linking profiles, porosity and permeability, biodegradability and biocompatibility, and sol-gel transitions [5]. Such work has paved the way for the translational research and clinical application of some of these polymers for drug delivery or to provide a scaffold for cell growth for the treatment of diabetes, arthritis, cancer and other ailments [6–17].

Among the conditions that may be ameliorated by the controlled, localized delivery of anti-proliferative drugs is vascular stenosis. Stenosis is a pathological narrowing of a blood vessel, typically a result of the migration and proliferation of vascular smooth muscle cells (SMC) and myofibroblasts into the neointimal layer. Stenosis is co-existent with large atherosclerotic lesions and restenosis often occurs after balloon angioplasty of such lesions [18]. In addition stenosis is prevalent after vein grafting [19], the placement of hemodialysis catheters, and native arteriovenous (AV) fistulas and synthetic grafts used for hemodialysis [20–22]. Research supports a role for adventitial fibroblasts in the stenotic disease that occurs after balloon angioplasty although the contribution of these cells may be model dependent [23–25]. Since perivascular cells (adventitial fibroblasts and SMC) participate in stenosis formation, and localized delivery circumvents undesirable side effects that can occur with systemic drug delivery, perivascular delivery of drug is a desirable means to combat vascular stenosis. And in fact, a number of methods have been explored for the perivascular delivery of anti-proliferative drugs to both the arterial and venous vasculature [26–36].

Our laboratory studies the pathogenesis and treatment of hemodialysis AV graft stenosis. AV grafts serve as a portal for hemodialysis, and stenosis within the graft can completely block blood flow or significantly impede high volume flow needed for reasonably short dialysis durations. The AV graft may be required for years in patients with end-stage renal disease due to the limited availability of kidneys for transplant. Unfortunately AV grafts fail at a high rate primarily due to stenosis development at the venous anastomosis [37, 38]. Drug-eluting stents and many perivascular delivery approaches are limited to a single application of drug at the time of graft implantation, and as such, may fail to inhibit stenosis that occurs at later time points. We suggest that a replenishable depot of drug may more effectively combat the deleterious cell proliferation that contributes to stenosis, and thereby indefinitely extend the time of graft patency.

ReGel™ is a triblock copolymer of poly(lactic-co-glycolic acid)-polyethylene glycol-poly(lactic-co-glycolic acid) (PLGA-PEG-PLGA). It is a thermosensitive hydrogel that is a liquid below 15 °C, but becomes a semisolid at body temperature. ReGel is a biodegradable polymer, which allows repeat administration and continuous drug delivery as it can be administered both at time of surgery, as well as percutaneously injected at later time points to replenish the depot.

We have previously reported the release profile and tissue distribution of anti-proliferative drugs (paclitaxel and dipyridamole) from different formulations of the polymer gel [26, 39]. Herein, we extend the characterization of ReGel-based drug delivery by examining the influence of drug loading methods on the in vitro drug release profiles of the hydrophobic anti-proliferative drug sirolimus (rapamycin), and report the in vivo tissue distribution of this drug. We also report the correlation of tissue drug concentrations determined post-mortem with the in vivo drug depot location as determined by non-invasive magnetic resonance imaging (MRI). These studies are part of a larger ongoing trial by our group examining the efficacy of sirolimus on inhibiting hyperplasia development in a porcine model of AV graft stenosis that will soon conclude.

2. Materials and Methods

2.1. Materials

Sirolimus (rapamycin) (>99% purity) and the internal standard everolimus (>99% purity) were purchased from LC Laboratories (Woburn, MA). The polymer gel, ReGel, was provided by Protherics, Salt Lake City, Inc., a BTG PLC Company, 1-chlorobutane, methanol, and acetonitrile were of HPLC grade, and were purchased from Sigma-Aldrich (St. Louis, MO). All other chemicals were of reagent grade, and were also purchased from Sigma-Aldrich. Expanded polytetrafluoroethylene (PTFE) grafts were purchased from Bard Peripheral Vascular Inc. (Temp, AZ).

2.2. In vitro drug release

The in vitro release profiles from ReGel of four formulations of sirolimus were examined. All in vitro experiments were performed in triplicate, and values were expressed as mean ± standard deviation. Samples were prepared by i) dissolving dry sirolimus in either methanol or dimethylsulfoxide (DMSO) (200 μL), and then mixing with five mL of liquid polymer on ice; or by ii) direct mixing of dry sirolimus powder (no organic solvent) with the liquid polymer at 4°C. The final concentration of all samples was 10 mg of sirolimus per mL of polymer. Drug-loaded liquid polymer samples (0.4 mL) were placed in 20 mL borosilicate scintillation vials and allowed to gel at 37°C. Release medium (5 mL), comprised of 1X PBS and 2% bovine serum albumin (BSA), was added to each vial on top of the polymer gel. To examine the effect of polymer on drug release, dry drug alone was incubated in release medium in the absence of polymer as well. Vials were incubated in a water bath at 37°C and agitated gently for the duration of the experiment. Release medium was removed daily from the sample vial and replaced with an equal volume of fresh medium. The release media samples for HPLC-MS/MS analysis were prepared by vortexing the sample (5 mL) with 1-chlorobutane (2 mL) for 2 min in 16 × 100 mm borosilicate test tubes. The tubes were then centrifuged at 25°C at 3500 rpm (2400 ×g) for 15 min. The organic layer was collected and solvent removed in vacuo at room temperature. The solid precipitate was then reconstituted in 200 μL of 1:1 H2O:CH3OH containing 0.1% formic acid and 2 mM ammonium acetate. The concentrations of sirolimus were determined using the HPLC MS/MS protocol described below.

2.3. Surgical Procedures

Animal procedures were performed according to protocols previously established in our laboratory [32]. These procedures conformed to the guidelines established by the Guidelines for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, revised 1996) and were approved by the Institutional Animal Care and Use Committee of the University of Utah and the Veterans Affairs Salt Lake City Healthcare System

A porcine AV PTFE graft model was employed as described in our previous studies [32]. In brief, Yorkshire domestic swine (~30 kg) were anesthetized by an intramuscular injection of a mixture of tiletamine/zolazepam (4 mg/kg; Fort Dodge Animal Health, Fort Dodge, IA), ketamine (4 mg/kg; Hospira Inc., Lake Forrest, IL), and xylazine (4 mg/kg; Lloyd Laboratories, Shenandoah, IA). General anesthesia was maintained using 1–3% isoflurane introduced via tracheal intubation. Intravenous sodium heparin (100 U/kg) was administered intraoperatively. Under sterile conditions, a spiral-reinforced expanded PTFE graft (7-cm length, 6-mm internal diameter) was placed between the common carotid artery and the ipsilateral external jugular vein (Figure 1). Post-operative analgesia was provided by intramuscular injection of buprenorphine hydrochloride (0.01 mg/kg; Reckitt Benckiser Healthcare Ltd., Hull, England) and application of a fentanyl patch (50 mg) (Watson Laboratories, Inc. Corona, CA).

Figure 1.

Figure 1

Open in a new tab

Panel a) Surgical placement of spiral-enforced polytetrafluoroethylene (PTFE) graft between the common carotid artery and the ipsilateral external jugular vein. Panel b) Polymer gel loaded with sirolimus placed at the graft-venous anastomosis intraoperatively.

Liquid polymer on ice and loaded with dry sirolimus (2.5 mg/mL in 2 mL of polymer) was applied to the venous anastomosis at the time of surgery immediately after graft placement (Figure 1). The gelling of the liquid polymer was visually confirmed before the surgical wound was closed. At one, two and three weeks after graft placement, additional sirolimus-loaded polymer was applied to the venous anastomosis by ultrasound guided injections as described below.

2.4 Ultrasound-guided injection of polymer

Ultrasound-guided injections of liquid drug-laden polymer were carried out according to previously described methods (Terry et al manuscript under review). Briefly, the animal was sedated with an intramuscular injection of a mixture of tiletamine/zolazepam, ketamine and xylazine as described above. The animal was placed in the supine position. The venous anastomosis was located using an ultrasound instrument and an L38 probe (Titan High Resolution Ultrasound System, SonoSite, Bothell, WA). Using sterile techniques, the cold liquid sirolimus-laden polymer was injected over approximately five seconds using a cold 18-gauge 102-mm-long Chiba-type needle (Hakko Co. Ltd, Tokyo, Japan) to the perivascular surface of the venous anastomosis. The animal was imaged by magnetic resonance imaging (MRI) within one hour of injection.

2.5. In vivo magnetic resonance imaging (MRI)

Animals were imaged by MRI as previously described (Terry et al manuscript under review). In brief, sedated animals were placed in a supine position within the bore of the MR scanner (3T Trio, Siemens Medical Solutions, Erlangen, Germany). The animals were imaged with a porcine-dedicated overlap-decoupled 16-channel phased array coil mounted on a fiberglass former previously molded to fit the porcine neck. The coil was placed directly over the surgical sites. A localization scan was performed followed by an axial 2D time-of-flight (TOF), with echo time (TE) of 5.9 ms and repetition time (TR) of 25 ms to verify proper positioning of the coil over the AV graft. A T2-weighted 3D turbo spin echo (TSE) sequence with restore pulse, and imaging parameters that were specifically optimized for the gel visualization (TE/TR=142/550 ms, echo train length (ETL)=17, voxel dimensions of 0.7mm×0.7mm×0.7mm), were applied to image the polymer gel. A T1-weighted 2D TSE black-blood sequence (TE/TR=8.6/800 ms, ETL=9, voxel dimensions of 2.0mm×0.6mm×0.6mm) was applied to visualize graft and blood vessel lumens.

2.6. In vivo drug distribution in tissues

The tissues from the porcine PTFE graft model were collected at various time points after polymer application and processed immediately after explantation. Tissue from one animal was used for each time point. The center point of the anastomosis was located, and carefully dissected away from the graft. The center point of vein, artery, and muscle were marked prior to explantation. After removal, each tissue was cut perpendicular to the longitudinal axis into 1-cm cross-sections both proximal and distal to the center of the anastomosis (Figure 2). The 1 cm-thick cross-sectional muscle segment was further separated into six pieces. The location (ventral vs. dorsal; left, right or central) of each section of the muscle segment was recorded in order to correlate with the drug concentrations. The position and weight of each tissue section were recorded.

Figure 2.

Figure 2

Open in a new tab

Cartoon of the explanted carotid-jugular graft and adjacent tissues. The tissues were cut transversely into multiple samples along the length and subjected to sirolimus extraction and analysis.

Regardless of whether or not gel was grossly visible on its surface, each tissue sample was scraped gently using a single-use stainless steel surgical blade and then rinsed with acetonitrile to remove any polymer gel or drug that might have attached to the surface of the tissue without penetration into the tissue [40]. The tissue sample was weighed (50 – 200 mg) then placed in a 1.5-mL microcentrifuge tube. A solution of 5% ZnSO4 (200 μL) and a solution of everolimus (500 ng/mL) in 1:1 H2O:CH3OH with 0.1% formic acid and 2 mM ammonium acetate (200 μL) were added. The everolimus served as an internal standard to monitor drug recovery during tissue extraction. Tissue samples were homogenized twice for 3 min each time at 25 Hz, using a high-throughput tissue disruptor (TissueLyser II, Qiagen, Valencia, CA). Acetonitrile (100 μL) was added to the homogenate, and the homogenized sample was then vortexed for 1 min. The resulting suspensions were centrifuged at 4000 rpm (1800 × g) for 10 min at room temperature. The supernatant was filtered using 4-mm diameter, 0.45-μm pore, PTFE syringe filters (Waters Corp., Milford, MA), and analyzed according to the HPLC MS/MS method described below.

2.7. High throughput sample analysis by HPLC tandem mass spectrometry

Sample analysis was adapted and modified from previously reported protocols [41–46]. The separation of sirolimus was performed using an XTerra C18 column (3.0 × 100 mm, 3.5 μm) (Waters Corp.) at 25° C on an Acquity 2695 HPLC system (Waters Corp.) equipped with a refrigerated autosampler. The mobile phase was composed of water and methanol (9:1 v/v) with 0.1% formic acid and 2 mM ammonium acetate. An isocratic method was employed with a flow rate set at 0.3 mL/min. The autosampler temperature was maintained at 20°C, and the sample injection volume was 30 μL. Sample run time was 6 min, followed by a 2-min needle wash process to prevent carryover. The internal standard eluted at 1.10 min, and sirolimus eluted at 1.17 min.

A triple quadrupole tandem mass spectrometer (Micromass® Quattro II, Waters Corp.) equipped with an electrospray ionization (ESI) interface was used for analytical detection. Quantification was performed in positive-ion mode using multiple reaction monitoring (MRM) of the transition masses of sirolimus (m/z 931.5 → 864.1) and the internal standard everolimus (m/z 975.5 → 908.1). Instrument parameters were optimized for the simultaneous detection of both the drug and internal standard. The capillary energy was set at 3.50 kV, cone voltage at 35 V, extractor at 3 V, source block temperature at 120° C, desolvation temperature at 400° C, and the optimal collision energy was determined to be 16 V. Calibration curves for both sirolimus and everolimus were established using standard samples at ten concentrations ranging from 0.5 μg/mL to 5 ng/mL. The coefficient of determination (r2) from a least squares linear regression was found to be 0.999 for sirolimus and 0.997 for everolimus. The lower limit of quantification (LLOQ) was 0.5 ng/mL for either compound.

The HPLC MS/MS method was validated by our lab for selectivity, accuracy, and consistency using spiked samples of release media, whole blood, venous tissue, arterial tissue, and muscular tissue. A >85% sample recovery was achieved, as determined by percent recovery of the internal standard, and no carryover was observed by inspection of blank injections in test runs. Corrections for sample loss were made by comparing the signal peak from the internal standard (everolimus 500 ng/ml) that was spiked in each experimental sample prior to extraction, with the signal peak from a calibration run using 500 ng/ml everolimus. The ratio of detected signal-to-predicted signal of the internal standard is directly proportional to the loss of sample, and was used to correct the signal intensity of individual sample runs.

2.8. MR image processing

OsiriX medical image processing software (Osirix v. 3.2.1) was used to examine the MR images and perform volume rendering. The graft/vessel lumens in 2D TSE black-blood sequences and the polymer gel in the 3D TSE sequences were selected using the grow-region tool. Signals from the tissues outside the selected regions of interest were set to below zero so that only the vessel/graft lumens and the polymer gel volumes were rendered. The lumens and the polymer gel regions of interest were then fused into one image and volume rendered using the 3D volume rendering tool.

2.9. Data Analysis

Calibration curve linear regression, graphing, and statistical significance calculations for in vitro drug release studies were performed using Graphpad Prism® software (La Jolla, CA). Individual one-tailed t-tests were performed to determine statistical significance between in vitro samples. A p value <0.05 is considered to be statistically significant.

3. Results and Discussion

3.1. In vitro drug release profiles

Initiating an initial burst in drug release followed by a more prolonged release profile could be an attractive means to inhibit both early and chronic events that contribute to hyperplasia formation. Thus, the in vitro release profile from polymer gel laden with undissolved “dry” drug, was compared to release profiles from polymer gel laden with drug dissolved in low concentrations of solvent (either methanol or DMSO). It was hypothesized that dissolution of sirolimus in solvent would promote an early release compared to the undissolved drug. The cumulative percentages of drug released from each of the four loading methods (methanol, DMSO, and dry sirolimus, all with polymer, or dry alone) are shown in Figure 3.

Figure 3.

Figure 3

Open in a new tab

Accumulation of sirolimus in release media in vitro as a percent of the initial amount loaded (10 mg/mL). A more rapid initial release was observed when sirolimus was dissolved in either DMSO or methanol before mixing with polymer (*p<0.05 vs. solid loading). At two weeks, however, there was no statistically significant difference in cumulative drug released from polymer among the various formulations.

Between two and 11 days, the amount of drug released from gel with drug dissolved in either methanol or DMSO prior to loading was significantly greater than that released from gel loaded with dry sirolimus (n = 3 each; p< 0.05). Likely, the difference in the higher release of both the methanol and DMSO formulations was due to the higher solubility of sirolimus as a result of the inclusion of organic solvents in the formulation, which allowed for the faster dissolution and diffusion of drug out of the polymer matrix. In contrast, the results suggest that dry loading of the hydrophobic sirolimus results in a stronger hydrophobic interaction between the drug and the polymer gel, and therefore a more gradual initial release. Dry drug in the absence of polymer gel only achieved ~13% accumulation in the media at 14 days, and accumulation was linear (r2= 0.988). This indicates the drug and polymer combination results in a higher solubility than dry drug alone (no polymer). This is likely due to alteration of the crystallinity of the drug by the polymer and/or formation of an inclusion complex between the amphiphilic polymer and the hydrophobic sirolimus that subsequently increased the drug solubility. It is also possible that the polymer may be forming micelles in the release media. Thus the polymer is not only a carrier suitable for injection of drug and a means by which to contain the drug at a specific location, but also enhances release of drug.

In spite of significant differences in the release profiles in the first 11 days between the formulations containing organic solvents (methanol or DMSO) and dry drug, there was no significant difference in the cumulative amount of drug released by day 14. It is likely that the drug release profile from the polymer depot is dependent on diffusion for the initial phase, after which drug release is dependent on both the hydrolysis of the polymer and on the diffusion of the drug. As such, each formulation apparently shares a similar release mechanism, but the organic solvents facilitate greater drug release in the initial phase. The release profile is typical of this and other thermosensitive hydrogels [2]. As such, the dry loading of sirolimus into the polymer gel was selected for use in the porcine PTFE graft model.

The results of the in vitro release studies also suggested the need for repeat dosing as 65–80% of the drug was released by day 14. For a six-week in vivo time course study, drug-loaded polymer may need to be readministered to the venous anastomosis approximately every week to two weeks to ensure a constant exposure of tissue to drug. However, even though the in vitro studies showed 80% of drug was released from the gel by two weeks, the drug may be retained in the tissues in vivo for a longer duration. Studies were therefore carried out in a porcine model to characterize the tissue pharmacokinetics.

3.2. In vivo drug concentration in porcine tissues

Drug concentrations were determined from porcine tissue samples using the protocol described in the Methods section at one, two, and six weeks after surgery and the results for venous and arterial wall tissue concentrations are summarized in Figure 4. Mean sirolimus concentrations in all tissue sections analyzed at one week were found to be below 3 ng/g of tissue (Figure 4a), and were lower at this early time point than the experimentally determined concentration that was required to inhibit the proliferation of cultured smooth muscle cells by 50% (IC50 = 5 ng/ml) [47].

Figure 4.

Figure 4

Open in a new tab

Sirolimus concentrations in venous and arterial tissues at various post-operative time points. Panel a) drug concentrations in tissue taken from one animal one week after graft placement. Two mL of polymer containing 5 mg of drug was applied to the graft-venous anastomosis at the time of surgery only; Panel b) Drug concentrations in tissue taken from one animal two weeks after graft placement. Two mL of polymer containing 5 mg of drug was applied to the graft-venous anastomosis at the time of surgery. At one week postoperatively, the drug depot was replenished by an ultrasound-guided injection of 2 mL of gel containing 5 mg of drug; Panel c) Drug concentrations in tissue taken from one animal six weeks after graft placement. The drug applications were the same as those described in panel b); in addition, a third injection of 2 mL of gel containing 5 mg of drug was applied at three weeks. This animal thus received 4 applications of polymer and drug. The error bars are derived from repeat analyses of the same extraction (each sample was analyzed twice) and by the propagation of error from the standard curve.

Our in vitro studies suggested that ~35% of drug should have been released from the polymer depot by one week (see Figure 3). Considering that a total of 5 mg of sirolimus (in 2 ml polymer) was applied to the venous anastomosis at time of surgery, approximately 1.75 mg of sirolimus is predicted to have been released from the gel for uptake into the tissue at this time point. The low drug concentrations in the walls throughout the AV graft and adjacent native vessels suggest that some of the drug might have been eliminated from the vascular tissues by metabolism or transport into the luminal blood stream, or might have diffused into the surrounding non-vascular tissues. Yet another potential explanation is that the drug depot might have migrated away from the anastomotic site post-operatively. The surgical placement of graft is an extensive surgical procedure that creates an open space around the graft-vessel anastomoses that may allow gel movements.

Additional experiments were performed in which the drug depot that was applied intra-operatively was replenished at various time points post-operatively. In these experiments, the sirolimus-laden polymeric gel was injected percutaneously under ultrasound guidance to the perivascular area around the venous anastomosis. In an animal that was euthanized at two weeks after graft placement, in which a drug depot was applied intra-operatively and then replenished at one week by ultrasound-guided injection, sirolimus concentrations were much higher in the wall of the AV graft and adjacent vascular tissues at two weeks (Figure 4b and Table 1) than those observed in tissue from an animal that was examined at one week post-operatively (Figure 4a). At two weeks, drug concentrations within 1 cm of the venous anastomosis (246 – 287 ng/mL) were 175 – 200-fold higher than at one week (1.4 – 1.5 ng/mL). Even greater concentrations of drug (>1500 ng/g) were found in tissue sections taken 2 cm proximal to the venous anastomosis indicating the gel had greater or longer contact with this section (Figure 4b).

Table 1.

Sirolimus concentrations in the juxta-anastomotic venous segments at various postoperative time points.

Venous tissue section location* 1 week 2 week 6 week
1 cm proximal to anastomosis 1.4 ng/g 246 ng/g 248 ng/g
1 cm distal to anastomosis 1.5 ng/g 287 ng/g 35 ng/g
Open in a new tab
*

refer to Figure 2 for explanation of the location

In another experiment, the drug depot applied intra-operatively was replenished at one week, two weeks and three weeks post-operatively by ultrasound-guided injection. The tissues were examined at six weeks after graft placement (three weeks after the last application of drug) and the results are shown in Figure 4c and Table 1. The highest concentration of drug was observed at the venous anastomosis. The drug was undetectable in the sections of vein and artery that are farther away from the anastomoses.

A total of 0.886 mg of sirolimus is accounted for in the two-week animal’s vascular tissues, which is only ~6% of the total 15 mg that were administered. It should be emphasized that these sirolimus concentrations only represent the spatial profile at a single time point. The potential explanations for the drug that could not be accounted for by this analysis have been noted above.

3.3. Correlation of tissue drug concentrations with in vivo images of polymer gel obtained by MRI

In other studies by our laboratory, MRI has proven useful to monitor polymer gel location in relationship to the AV graft (Terry et al manuscript, under review). For the current study, MRI was performed on an animal that had drug-laden polymer gel applied to the venous anastomoses of the AV graft at time of surgery and had received an ultrasound-guided injection of a replenishment of drug-laden polymer one week later. The AV graft was placed on the left side of the animal. Polymer was also injected near the contralateral unoperated jugular vein on the right side of the same animal at both time of surgery and one week later. Injection of drug on the unoperated side of the animal was undertaken to obtain images and drug concentrations from a region that did not have a surgical wound that could allow for the migration of the polymeric gel. Five days after the injection of polymeric gel to the unoperated region and to the contralateral AV graft venous anastomosis (a total of twelve days after surgery), the animal was imaged by MRI and immediately euthanized. The AV graft with the adjacent tissues and the unoperated contralateral vein were explanted. The tissue drug concentrations in the respective explants were then compared with the in vivo images of each polymer depot as shown in Figure 5 (AV graft) and Figure 6 (unoperated contralateral vein).

Figure 5.

Figure 5

Open in a new tab

Three-dimensional (3D) reconstruction of a carotid-jugular graft and polymer gel from MRI and correlation to tissue drug concentrations. The polymer gel was applied to the graft-venous anastomosis at time of graft placement and then injected to the same region one week later. MRI was performed five days after the injection (12 days after graft placement). The polymer was reconstructed in pale green, while the blood vessels and graft were reconstructed in red, from the MR image. The sternocleidomastoid muscle was not included in the 3D reconstruction. Panel a) Dorsal view of the 3D reconstruction of the graft and polymer gel. Panel b) Ventral view of the same 3D reconstruction of the graft and polymer gel shown in panel a). Panel c) Graphic representation of measured sirolimus concentrations in various segments of tissues collected from the same animal described in panel a) and panel b). Each color bar in the linear concentration scale represents a 187.5 ng/g increment in concentration.

Figure 6.

Figure 6

Open in a new tab

Three-dimensional (3D) reconstruction of polymer gel adjacent to external jugular vein without graft from MRI and correlation to tissue drug concentrations. The polymer was injected percutaneously to the perivascular area adjacent to the contralateral external jugular vein without graft at the time of graft placement on one side. The injection of polymer to the same region was repeated one week later. MRI was performed five days after the last injection (12 days after graft placement surgery). The polymer was reconstructed in pale green, while the blood vessels and graft were reconstructed in red, from the MR image. The sternocleidomastoid muscle was not included in the 3D reconstruction. Panel a) Dorsal view of the 3D reconstruction of the external jugular vein, adjacent polymer gel and common carotid artery. Panel b) Ventral view of the same 3D reconstruction of vein, polymer gel and artery shown in panel a). Panel c) Graphic representation of measured sirolimus concentrations in various segments of tissues collected from the same animal described in panel a) and panel b). Each color bar in the linear concentration scale represents a 187.5 ng/g increment in concentration.

In both cases, the highest levels of sirolimus were observed adjacent to the drug depot as imaged in vivo. The drug concentrations in the vascular walls obtained approximately three weeks after graft placement were much higher than the drug’s effective in vitro anti-proliferative concentrations, with the highest concentrations found around the gel depot (venous anastomosis) and adjacent sections, as confirmed by the MR images. The arterial anastomosis of the AV graft showed higher drug concentrations than the section of muscle that separated the arterial from the venous anastomoses (Figure 5c). This suggests that the polymer might have migrated around but not through the muscle tissue towards the arterial anastomosis. This presumption appears to be confirmed by the MR image shown in Figure 5(a) and (b). In contrast, Figure 6(c) shows that drug concentrations on the right side of the neck that did not receive graft placement were high only in the venous wall adjacent to the drug depot. The muscle tissues contained relatively little drug.

The distribution of drug concentrations is not homogenous throughout the vascular tissues or the muscle tissues, as shown in Figure 5 and Figure 6. Thus, detailed studies of tissue pharmacokinetics suggest that multiple small sections should be obtained to provide a higher resolution picture of drug distribution in tissue.

The protocol of applying drug to the venous anastomosis a total of four times in a three week period resulted in drug concentrations that were substantially higher than the concentrations determined to be effective against smooth muscle cell proliferation in in vitro studies. Thus lower drug concentrations per application or fewer application times may be sufficient to maintain local drug concentrations that are effective against hyperplasia formation in vivo. Studies from our group are completed that indicate the efficacy of the current drug administration protocol on inhibiting hyperplasia formation in the porcine AV graft model. Other studies can now be carried out to determine the dosing strategy that provides the minimal concentration that is still effective. It may be that much higher levels of drug are required in vivo than what is effective in in vitro assays.

4. Conclusions

The local perivascular sustained delivery of anti-proliferative drugs to the target site using injectable polymeric gel is a logical and attractive strategy for the prevention and treatment of neointimal hyperplasia. In the present study, we found that a single application of sirolimus-loaded polymer at the time of AV graft placement surgery in the porcine model may not provide an effective drug depot, partly because of the potential migration of the gel. However, repeated ultrasound-guided injections of drug-laden gel to replenish the depot, that exploit the thermosensitive and biodegradation property of the polymer, resulted in a build up of vascular tissue sirolimus concentrations that were substantially higher than the drug’s effective in vitro anti-proliferative concentrations. This study has confirmed that MR images of injected polymer gel correlate well with ex vivo tissue drug concentrations. Thus, MR imaging of the polymer gel location is currently being used to assist in the evaluation of results of a larger trial being carried out by our group to study the efficacy of sirolimus on inhibiting hyperplasia development in our porcine model of AV graft stenosis. A manuscript reporting the results of that study is in preparation.

Acknowledgments

This work was supported by RO1HL67646 from the National Heart, Lung and Blood Institute and the Department of Veterans Affairs.

We would like to thank Ramesh Rathi and Kirk Fowers of Protherics, Salt Lake City, Inc., a BTG PLC Company, as well as Jim Muller of the University of Utah, Department of Chemistry Mass Spectrometry Facility for their technical support and advice. We would also like to thank Ilya Zhuplatov for his assistance during surgical procedures.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Contributor Information

Shawn C. Owen, Department of Pharmaceutics and Pharmaceutical Chemistry, College of Pharmacy, 301 Skaggs Hall, 30 South 2000 East, Salt Lake City, UT 84112

Huan Li, Division of Nephrology and Hypertension, University of Utah, 85 N. Medical Dr. East, Salt Lake City, UT 84112

William G. Sanders, Department of Pharmaceutics and Pharmaceutical Chemistry, College of Pharmacy, 301 Skaggs Hall, 30 South 2000 East, Salt Lake City, UT 84112

Alfred K. Cheung, Medical Service, Veterans Affairs Salt Lake City Healthcare System, 500 Foothill Dr., Salt Lake City, UT 84148, Division of Nephrology and Hypertension, University of Utah, 85 N. Medical Dr. East, Salt Lake City, UT 84112

Christi M. Terry, Email: Christi.terry@hsc.utah.edu, Division of Nephrology and Hypertension, University of Utah, 85 N. Medical Dr. East, Salt Lake City, UT 84112, Tele: 801-582-1565 ext 5380, Fax: 801-524-5620

References

  • 1.Hoffman AS. Hydrogels for biomedical applications. Adv Drug Deliv Rev. 2002;54(1):3–12. doi: 10.1016/s0169-409x(01)00239-3. [DOI] [PubMed] [Google Scholar]
  • 2.Jeong B, Kim SW, Bae YH. Thermosensitive sol-gel reversible hydrogels. Adv Drug Deliv Rev. 2002;54(1):37–51. doi: 10.1016/s0169-409x(01)00242-3. [DOI] [PubMed] [Google Scholar]
  • 3.He C, Kim SW, Lee DS. In situ gelling stimuli-sensitive block copolymer hydrogels for drug delivery. J Control Release. 2008;127(3):189–207. doi: 10.1016/j.jconrel.2008.01.005. [DOI] [PubMed] [Google Scholar]
  • 4.Zentner GM, Rathi R, Shih C, McRea JC, Seo MH, Oh H, Rhee BG, Mestecky J, Moldoveanu Z, Morgan M, Weitman S. Biodegradable block copolymers for delivery of proteins and water-insoluble drugs. J Control Release. 2001;72(1–3):203–215. doi: 10.1016/s0168-3659(01)00276-0. [DOI] [PubMed] [Google Scholar]
  • 5.Uchegbu IF, Schätzlein AG. Polymers in drug delivery. CRC/Taylor & Francis; Boca Raton, FL: 2006. p. 259. [Google Scholar]
  • 6.Bergman K, Engstrand T, Hilborn J, Ossipov D, Piskounova S, Bowden T. Injectable cell-free template for bone-tissue formation. J Biomed Mater Res A. 2008 doi: 10.1002/jbm.a.32289. [DOI] [PubMed] [Google Scholar]
  • 7.Park JS, Woo DG, Yang HN, Na K, Park KH. Transforming growth factor beta-3 bound with sulfate polysaccharide in synthetic extracellular matrix enhanced the biological activities for neocartilage formation in vivo. J Biomed Mater Res A. 2008 doi: 10.1002/jbm.a.32271. [DOI] [PubMed] [Google Scholar]
  • 8.Yang KC, Wu CC, Cheng YH, Kuo TF, Lin FH. Chitosan/gelatin hydrogel prolonged the function of insulinoma/agarose microspheres in vivo during xenogenic transplantation. Transplant Proc. 2008;40(10):3623–3626. doi: 10.1016/j.transproceed.2008.06.092. [DOI] [PubMed] [Google Scholar]
  • 9.Yang KC, Wu CC, Lin FH, Qi Z, Kuo TF, Cheng YH, Chen MP, Sumi S. Chitosan/gelatin hydrogel as immunoisolative matrix for injectable bioartificial pancreas. Xenotransplantation. 2008;15(6):407–416. doi: 10.1111/j.1399-3089.2008.00503.x. [DOI] [PubMed] [Google Scholar]
  • 10.Huynh DP, Im GJ, Chae SY, Lee KC, Lee DS. Controlled release of insulin from pH/temperature-sensitive injectable pentablock copolymer hydrogel. J Control Release. 2009 doi: 10.1016/j.jconrel.2009.02.021. [DOI] [PubMed] [Google Scholar]
  • 11.Kim HK, Shim WS, Kim SE, Lee KH, Kang E, Kim JH, Kim K, Kwon IC, Lee DS. Injectable in situ-forming pH/thermo-sensitive hydrogel for bone tissue engineering. Tissue Eng Part A. 2009;15(4):923–933. doi: 10.1089/ten.tea.2007.0407. [DOI] [PubMed] [Google Scholar]
  • 12.Lee J, Lee KY. Local and sustained vascular endothelial growth factor delivery for angiogenesis using an injectable system. Pharm Res. 2009;26(7):1739–1744. doi: 10.1007/s11095-009-9884-4. [DOI] [PubMed] [Google Scholar]
  • 13.Saito M, Takahashi KA, Arai Y, Inoue A, Sakao K, Tonomura H, Honjo K, Nakagawa S, Inoue H, Tabata Y, Kubo T. Intraarticular administration of platelet-rich plasma with biodegradable gelatin hydrogel microspheres prevents osteoarthritis progression in the rabbit knee. Clin Exp Rheumatol. 2009;27(2):201–207. [PubMed] [Google Scholar]
  • 14.Swindle KE, Shah M, Hamilton PD, Eskin TA, Kaushal S, Ravi N. Rabbit study of an in situ forming hydrogel vitreous substitute. Invest Ophthalmol Vis Sci. 2009 doi: 10.1167/iovs.08-2891. [DOI] [PubMed] [Google Scholar]
  • 15.Vukelja SJ, Anthony SP, Arseneau JC, Berman BS, Cunningham CC, Nemunaitis JJ, Samlowski WE, Fowers KD. Phase 1 study of escalating-dose OncoGel (ReGel/paclitaxel) depot injection, a controlled-release formulation of paclitaxel, for local management of superficial solid tumor lesions. Anticancer Drugs. 2007;18(3):283–289. doi: 10.1097/CAD.0b013e328011a51d. [DOI] [PubMed] [Google Scholar]
  • 16.DuVall GA, Tarabar D, Seidel RH, Elstad NL, Fowers KD. Phase 2: a dose-escalation study of OncoGel (ReGel/paclitaxel), a controlled-release formulation of paclitaxel, as adjunctive local therapy to external-beam radiation in patients with inoperable esophageal cancer. Anticancer Drugs. 2009;20(2):89–95. doi: 10.1097/CAD.0b013e3283222c12. [DOI] [PubMed] [Google Scholar]
  • 17.Elstad NL, Fowers KD. OncoGel (ReGel/paclitaxel) - Clinical applications for a novel paclitaxel delivery system. Adv Drug Deliv Rev. 2009 doi: 10.1016/j.addr.2009.04.010. [DOI] [PubMed] [Google Scholar]
  • 18.Schillinger M, Minar E. Restenosis after percutaneous angioplasty: the role of vascular inflammation. Vasc Health Risk Manag. 2005;1(1):73–78. doi: 10.2147/vhrm.1.1.73.58932. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Kalan JM, Roberts WC. Morphologic findings in saphenous veins used as coronary arterial bypass conduits for longer than 1 year: necropsy analysis of 53 patients, 123 saphenous veins, and 1865 five-millimeter segments of veins. Am Heart J. 1990;119(5):1164–1184. doi: 10.1016/s0002-8703(05)80249-2. [DOI] [PubMed] [Google Scholar]
  • 20.Li L, Terry CM, Shiu YT, Cheung AK. Neointimal hyperplasia associated with synthetic hemodialysis grafts. Kidney Int. 2008;74(10):1247–1261. doi: 10.1038/ki.2008.318. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Wang Y, Krishnamoorthy M, Banerjee R, Zhang J, Rudich S, Holland C, Arend L, Roy-Chaudhury P. Venous stenosis in a pig arteriovenous fistula model--anatomy, mechanisms and cellular phenotypes. Nephrol Dial Transplant. 2008;23(2):525–533. doi: 10.1093/ndt/gfm547. [DOI] [PubMed] [Google Scholar]
  • 22.Yevzlin AS. Hemodialysis catheter-associated central venous stenosis. Semin Dial. 2008;21(6):522–527. doi: 10.1111/j.1525-139X.2008.00496.x. [DOI] [PubMed] [Google Scholar]
  • 23.Shi Y, O’Brien JE, Fard A, Mannion JD, Wang D, Zalewski A. Adventitial myofibroblasts contribute to neointimal formation in injured porcine coronary arteries. Circulation. 1996;94(7):1655–1664. doi: 10.1161/01.cir.94.7.1655. [DOI] [PubMed] [Google Scholar]
  • 24.Scott NA, Cipolla GD, Ross CE, Dunn B, Martin FH, Simonet L, Wilcox JN. Identification of a potential role for the adventitia in vascular lesion formation after balloon overstretch injury of porcine coronary arteries. Circulation. 1996;93(12):2178–2187. doi: 10.1161/01.cir.93.12.2178. [DOI] [PubMed] [Google Scholar]
  • 25.De Leon H, Ollerenshaw JD, Griendling KK, Wilcox JN. Adventitial cells do not contribute to neointimal mass after balloon angioplasty of the rat common carotid artery. Circulation. 2001;104(14):1591–1593. [PubMed] [Google Scholar]
  • 26.Masaki T, Rathi R, Zentner G, Leypoldt JK, Mohammad SF, Burns GL, Li L, Zhuplatov S, Chirananthavat T, Kim SJ, Kern S, Holman J, Kim SW, Cheung AK. Inhibition of neointimal hyperplasia in vascular grafts by sustained perivascular delivery of paclitaxel. Kidney Int. 2004;66(5):2061–2069. doi: 10.1111/j.1523-1755.2004.00985.x. [DOI] [PubMed] [Google Scholar]
  • 27.Lipke EA, West JL. Localized delivery of nitric oxide from hydrogels inhibits neointima formation in a rat carotid balloon injury model. Acta Biomater. 2005;1(6):597–606. doi: 10.1016/j.actbio.2005.07.010. [DOI] [PubMed] [Google Scholar]
  • 28.Pires NM, Schepers A, van der Hoeven BL, de Vries MR, Boesten LS, Jukema JW, Quax PH. Histopathologic alterations following local delivery of dexamethasone to inhibit restenosis in murine arteries. Cardiovasc Res. 2005;68(3):415–424. doi: 10.1016/j.cardiores.2005.06.015. [DOI] [PubMed] [Google Scholar]
  • 29.Pires NM, van der Hoeven BL, de Vries MR, Havekes LM, van Vlijmen BJ, Hennink WE, Quax PH, Jukema JW. Local perivascular delivery of anti-restenotic agents from a drug-eluting poly(epsilon-caprolactone) stent cuff. Biomaterials. 2005;26(26):5386–5394. doi: 10.1016/j.biomaterials.2005.01.063. [DOI] [PubMed] [Google Scholar]
  • 30.Banno H, Takei Y, Muramatsu T, Komori K, Kadomatsu K. Controlled release of small interfering RNA targeting midkine attenuates intimal hyperplasia in vein grafts. J Vasc Surg. 2006;44(3):633–641. doi: 10.1016/j.jvs.2006.04.044. [DOI] [PubMed] [Google Scholar]
  • 31.Goteti K, Masaki T, Kuji T, Leypoldt JK, Cheung AK, Kern SE. Perivascular tissue pharmacokinetics of dipyridamole. Pharm Res. 2006;23(4):718–728. doi: 10.1007/s11095-006-9746-2. [DOI] [PubMed] [Google Scholar]
  • 32.Kuji T, Masaki T, Goteti K, Li L, Zhuplatov S, Terry CM, Zhu W, Leypoldt JK, Rathi R, Blumenthal DK, Kern SE, Cheung AK. Efficacy of local dipyridamole therapy in a porcine model of arteriovenous graft stenosis. Kidney Int. 2006;69(12):2179–2185. doi: 10.1038/sj.ki.5000383. [DOI] [PubMed] [Google Scholar]
  • 33.Melhem M, Kelly B, Zhang J, Kasting G, Li J, Davis H, Heffelfinger S, Desai P, Roy-Chaudhury P. Development of a local perivascular paclitaxel delivery system for hemodialysis vascular access dysfunction: polymer preparation and in vitro activity. Blood Purif. 2006;24(3):289–298. doi: 10.1159/000091346. [DOI] [PubMed] [Google Scholar]
  • 34.Pires NM, Eefting D, de Vries MR, Quax PH, Jukema JW. Sirolimus and paclitaxel provoke different vascular pathological responses after local delivery in a murine model for restenosis on underlying atherosclerotic arteries. Heart. 2007;93(8):922–927. doi: 10.1136/hrt.2006.102244. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.El-Kurdi MS, Hong Y, Stankus JJ, Soletti L, Wagner WR, Vorp DA. Transient elastic support for vein grafts using a constricting microfibrillar polymer wrap. Biomaterials. 2008;29(22):3213–3220. doi: 10.1016/j.biomaterials.2008.04.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Pearce CG, Najjar SF, Kapadia MR, Murar J, Eng J, Lyle B, Aalami OO, Jiang Q, Hrabie JA, Saavedra JE, Keefer LK, Kibbe MR. Beneficial effect of a short-acting NO donor for the prevention of neointimal hyperplasia. Free Radic Biol Med. 2008;44(1):73–81. doi: 10.1016/j.freeradbiomed.2007.09.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Etheredge EE, Haid SD, Maeser MN, Sicard GA, Anderson CB. Salvage operations for malfunctioning polytetrafluoroethylene hemodialysis access grafts. Surgery. 1983;94(3):464–470. [PubMed] [Google Scholar]
  • 38.Palder SB, Kirkman RL, Whittemore AD, Hakim RM, Lazarus JM, Tilney NL. Vascular access for hemodialysis. Patency rates and results of revision. Ann Surg. 1985;202(2):235–239. doi: 10.1097/00000658-198508000-00015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Zhu W, Masaki T, Bae YH, Rathi R, Cheung AK, Kern SE. Development of a sustained-release system for perivascular delivery of dipyridamole. J Biomed Mater Res B Appl Biomater. 2006;77(1):135–143. doi: 10.1002/jbm.b.30412. [DOI] [PubMed] [Google Scholar]
  • 40.Matthes K, Mino-Kenudson M, Sahani DV, Holalkere N, Fowers KD, Rathi R, Brugge WR. EUS-guided injection of paclitaxel (OncoGel) provides therapeutic drug concentrations in the porcine pancreas (with video) Gastrointest Endosc. 2007;65(3):448–453. doi: 10.1016/j.gie.2006.06.030. [DOI] [PubMed] [Google Scholar]
  • 41.Kahan BD, Napoli KL, Kelly PA, Podbielski J, Hussein I, Urbauer DL, Katz SH, Van Buren CT. Therapeutic drug monitoring of sirolimus: correlations with efficacy and toxicity. Clin Transplant. 2000;14(2):97–109. doi: 10.1034/j.1399-0012.2000.140201.x. [DOI] [PubMed] [Google Scholar]
  • 42.Taylor PJ, Salm P, Lynch SV, Pillans PI. Simultaneous quantification of tacrolimus and sirolimus, in human blood, by high-performance liquid chromatography-tandem mass spectrometry. Ther Drug Monit. 2000;22(5):608–612. doi: 10.1097/00007691-200010000-00017. [DOI] [PubMed] [Google Scholar]
  • 43.Streit F, Armstrong VW, Oellerich M. Rapid liquid chromatography-tandem mass spectrometry routine method for simultaneous determination of sirolimus, everolimus, tacrolimus, and cyclosporin A in whole blood. Clin Chem. 2002;48(6 Pt 1):955–958. [PubMed] [Google Scholar]
  • 44.Boernsen KO, Egge-Jacobsen W, Inverardi B, Strom T, Streit F, Schiebel HM, Benet LZ, Christians U. Assessment and validation of the MS/MS fragmentation patterns of the macrolide immunosuppressant everolimus. J Mass Spectrom. 2007;42(6):793–802. doi: 10.1002/jms.1215. [DOI] [PubMed] [Google Scholar]
  • 45.Wang S, Miller A. A rapid liquid chromatography-tandem mass spectrometry analysis of whole blood sirolimus using turbulent flow technology for online extraction. Clin Chem Lab Med. 2008;46(11):1631–1634. doi: 10.1515/CCLM.2008.303. [DOI] [PubMed] [Google Scholar]
  • 46.Seger C, Tentschert K, Stoggl W, Griesmacher A, Ramsay SL. A rapid HPLC-MS/MS method for the simultaneous quantification of cyclosporine A, tacrolimus, sirolimus and everolimus in human blood samples. Nat Protoc. 2009;4(4):526–534. doi: 10.1038/nprot.2009.25. [DOI] [PubMed] [Google Scholar]
  • 47.Zhu W. Sustained perivascular drug delivery for the prevention of hemodialysis arteriovenous graft stenosis. Dept., of Pharmaceutics and Pharmaceutical Chemistry, University of Utah; 2006. p. xiv. 156 leaves. [Google Scholar]

RESOURCES