Preclinical Evaluation of Second-Generation Everolimus- and Zotarolimus-Eluting Coronary Stents

Abstract: Objectives. This study was designed to evaluate the pharmacokinetic and vascular healing of a second-generation everolimus-eluting stent (EES) and slow-release zotarolimus-eluting stent (R-ZES). Background. Second-generation DESs have alleviated the safety concerns of late stent thrombosis by addressing issues of polymer biocompatibility and stent design, and optimizing drug loads and release kinetics. No preclinical comparison study exists between these stents. Methods. Rabbit iliac artery stent implantation was performed using Xience Prime EES and Resolute R-ZES. Histomorphometric evaluation was performed at 28 and 60 days after implantation in an induced atheroma model. Endothelial coverage and maturation were assessed by scanning electron microscopy and immuno-labeling at 14 and 28 days following deployment. For pharmacokinetic studies, arterial tissue and stents were retrieved at 3, 14, 28, and 90 days, and blood samples were obtained during the first 24 hours. Results. Vascular remodeling (percent stenosis, neointimal thickness) was similar in arteries implanted with either stent group. At 28 days, inflammation was significantly less in the EES group as compared to the R-ZES group (inflammation score: 1.59 ± 0.52 vs 2.22 ± 0.69, respectively; P=.044), with no differences observed at 60 days. Endothelial coverage was similar between both groups; however, endothelial maturation above stent struts was significantly higher in the EES group vs R-ZES group at 28 days (33 ± 20% vs 22 ± 21%, respectively; P=.040). Arterial drug level concentrations were also shown to be significantly less in the EES group vs the R-ZES group (P<.0001). Conclusions. Overall, EES and R-ZES displayed similar remodeling properties with lower arterial drug levels observed in the EES group vs the R-ZES group, which may have led to more rapid endothelial maturation.

J INVASIVE CARDIOL 2013;25(8):383-390

Key words: everolimus-eluting stent, zotarolimus-eluting stent, rabbit iliac model, histopathology, pharmacokinetics

____________________________________

First-generation drug eluting stents (DESs) showed superior results in reducing restenosis rates as compared to bare metal stents (BMSs); however, late stent thrombosis (LST) emerged as a safety concern.^1,2 Newer DESs have approached concerns of LST by addressing issues of polymer biocompatibility and stent design, and reducing/optimizing drug loads and release kinetics. Though different design approaches are under clinical evaluation to address concerns of LST, including for example the application of biodegradable polymer coatings, biodegradable scaffolds, and drug-coated balloons, the purported benefits of these concepts remains to be demonstrated. 

Evaluating the impact of varying design changes can be challenging, as DES healing involves multiple biological responses, starting with stent-induced injury, thrombus formation, inflammation, and smooth muscle cell and endothelial cell proliferation.^3,4 Preclinical models help to discern differences in devices as the sequence of biological events associated with arterial repair is similar to humans.⁵ The rabbit ilio-femoral artery model has been shown to be an effective model in evaluating vascular healing following DES implantation, in particular to assess reendothelialization rates.⁶ Recent studies have suggested that atherosclerotic models could serve as better models to discern differences between next-generation devices.⁷ 

The objective of this study was to evaluate the pharmacokinetic and vascular healing characteristics of two new-generation DESs. The Xience Prime (Abbott Vascular) elutes everolimus at a dose of 100 µg/cm² via a fluorinated copolymer on a cobalt-chromium Multi-Link stent. The Endeavor Resolute (Medtronic) elutes zotarolimus at a dose of 160 µg/cm² via a proprietary blended polymer of a hydrophilic and hydrophobic polymer and water-soluble polyvinyl pyrrolidone on a cobalt-nickel Driver stent. These two stents were compared head-to-head using both healthy and atherosclerotic rabbit ilio-femoral artery models to compare the drug release and arterial drug retention, reendothelialization, and vascular healing. 

Methods

This study was approved by the Institutional Animal Care and Use Committee and conformed to the position of the American Heart Association on use of animals in research. In this series of experiments, the vascular responses to stents were compared assessing healing and inflammation with histology and immunohistochemistry in a rabbit model of experimental atherosclerosis. Rates of endothelialization and pharmacokinetic (PK) profiles were also evaluated in a non-diseased rabbit model using established methodologies. Time points for all assessments were chosen based upon the model, experimental variable being assessed, and published literature. The data from these three experiments are compiled herein to present a more complete picture of the in vivo comparison of the two devices. Figure 1 summarizes the number of animals and stents used in this study. 

Rabbit model of experimental atherosclerosis. The experimental preparation of the atherosclerotic animal model has been previously reported.^7,8 Briefly, New Zealand White rabbits (3 to 4 kg) 3 to 4 months of age were fed an atherogenic diet (1% cholesterol and 6% peanut oil; F4366-CHL, Bio-Serv, Inc) for 5 weeks to induce atherosclerosis. Iliac artery injury was induced 1 week following induction of a high-cholesterol diet using a Fogarty catheter (3 Fr). Following balloon injury, the animals were maintained on an atherogenic diet for 4 weeks. Subsequently, the diet was switched to a low-cholesterol diet (containing 0.025% cholesterol) for the remainder of the study.

Stent placement and harvest. Under fluoroscopic guidance, anesthetized adult male New Zealand White rabbits underwent endothelial denudation (non-athero model only) of both iliac arteries using an angioplasty balloon catheter (3.0 x 10 mm; Abbott Vascular). Subsequently, premounted Xience Prime EESs (3 x 12 mm) or Endeavor Resolute R-ZESs (3 x 12 mm) were implanted over lesions in both iliac arteries using respective nominal pressures to achieve a visual target stent-to-artery ratio of approximately 1.3:1. For evaluations using histologic techniques, stents were implanted in contralateral vessels to enable paired comparison between stents and minimize animal-to-animal variability. For PKs, the same DES was used for a given animal to avoid cross-drug contamination of the analytical assay. After stent implantation, postprocedural angiography was performed to document vessel patency, and the animals were allowed to recover. Antiplatelet therapy consisted of aspirin (40 mg/day) given orally 24 hours before catheterization with continued dosing throughout the in-life phase of the study, while single-dose intraarterial heparin (150 IU/kg) was administered at the time of catheterization. At the designated time point, animals were anesthetized followed by euthanasia and perfusion-fixation. 

Light microscopy and histomorphometry. Vessels from the 28- and 60-day atherosclerotic animals were used for histologic assessment of healing and inflammation. The samples were further fixed by immersion and then embedded in methylmethacrylate for light microscopy analysis with sections taken from the proximal, middle, and distal portions of the stent. Histologic sections were stained with H&E and Movat’s Pentachrome. Additional sections from the mid-portion of the implanted arterial segments from the atherosclerotic rabbit ilio-femoral arteries were held in reserve for immunohistochemical identification of macrophages (RAM11). RAM11 is a mouse monoclonal IgG1 anti-rabbit macrophage antibody, which is widely used for immunohistochemical demonstration of rabbit macrophages, particularly in the cellular analysis of the atherosclerotic lesions of cholesterol-fed rabbits as well as in morphological analysis of normal rabbit arteries. Histological sections were measured using computer-assist software (IPLab; Scanalytics, BD Biosciences). Cross-sectional area measurements included the external elastic lamina, stent, and lumen. A mean neointimal thickness was determined as the distance at the inner strut surface to the luminal border. In addition to area and thickness measurements, stented segments were assessed for the presence of uncovered struts and fibrin.⁷ Inflammation was evaluated by the following score system: Score 0 = <25% struts with fewer than 10 inflammatory cells; Score 1 = up to 25% struts with greater than 10 inflammatory cells; Score 2 = 25%-50% struts with greater than 10 inflammatory cells; Score 3 = >50% struts with greater than 10 inflammatory cells; Score 4 = ≥2  strut-associated granulomatous inflammatory reactions. 

En face assessment of endothelialization. Assessment of endothelialization was performed in normal rabbits using the ilio-femoral artery model as previously described.⁹ As previously established, 14 and 28 days represent time points wherein differentiation between DESs can be observed. At 14 and 28 days, animals were euthanized and vessels were perfusion-fixed. The samples were further fixed by immersion and then bisected longitudinally for scanning electron microscopy (SEM) and confocal analysis. Composites of serial en face SEM images were acquired at low power (15x magnification) and digitally assembled to provide a complete view of the entire luminal stent surface. The images were further enlarged (200x magnification), allowing direct visualization of endothelial cells. Immunostaining of whole-mount specimens was achieved by overnight incubation in an antibody cocktail containing primary antibodies against PECAM-1 (CD31, 1:20 dilution; Dako) and thrombomodulin (TM) (1:10 dilution; American Diagnostica) at 4 °C. Platelet endothelial cell adhesion molecule is an endothelial cell-cell junction marker used to demonstrate the presence of endothelial cells in histological tissue sections. The extent of endothelial coverage based on a positive reaction to PECAM-1 above and between stent struts was visually estimated for each level of struts and expressed as a mean percentage for the total area above and between stent struts for the entire stented surface. TM is a membrane-bound protein expressed on endothelial cells that reduces blood coagulation. TM expression was assessed based on reaction intensity ranging from an absence to strong staining relative to non-adjacent control segments. 

Pharmacokinetic evaluations. For the PK evaluations, EESs and R-ZESs were explanted at 3, 14, 28, and 90 days after deployment using the normal rabbit ilio-femoral model. Non-diseased animals are recommended by the Food and Drug Administration of the United States for PK evaluation of DESs.¹⁰ Additionally, non-diseased rabbits were utilized to minimize the variability on drug/tissue concentrations due to varying amounts of atheroma. In parallel, blood samples were obtained at 30 minutes, 3 hours and 24 hours after treatment. Tissue drug concentration measurements of the artery and circulating plasma were performed by liquid chromatography-mass spectrometry (LCMS). Drug concentrations of the stent were performed by reversed-phase high-performance liquid chromatography (RP-HPLC), as described previously.¹¹ 

Statistical analysis. Continuous data were expressed as mean ± standard deviation. Normality of distribution was tested with the Wilk-Shapiro test (JMP software). Statistical comparisons of continuous data were performed by one-way analysis of variance (ANOVA). Non-parametric data were examined by Wilcoxon rank-sum test to assess significant differences. For arterial PK evaluation, the area under the curve (AUC) method was utilized.¹² A P-value of ≤.05 was considered statistically significant.

Results

Animal condition. In the atherosclerotic rabbits, two animals were excluded due to low response to the cholesterol diet. A third animal was euthanized prior to the day 35 diet change due to poor health (anorexia). During stenting, one animal was found to have bilateral angiographic occlusion of the iliac arteries before stent implantation and was subsequently excluded from the implantation matrix. Thirty-six stents were harvested from 18 animals, with an equal number of EESs and R-ZESs (n = 9) for each group for the 28- and 60-day time points. Mean levels of circulating cholesterol are provided in Supplemental Table 1.

In the healthy rabbits, all stents were successfully deployed without dissection or thrombosis and all animals remained alive for the duration of the study. 

Pharmacokinetics. The PK results are summarized in Table 1. A steady decrease in drug concentration was shown in arterial tissue for both stents, although in general, R-ZES implanted arteries contained a 10-fold greater drug content than EES implanted arteries (3 days: 1.89 vs 11.9 ng/mg; 14 days: 0.91 vs 12.7 ng/mg; 28 days: 0.84 vs 10.8 ng/mg; 90 days: 0.38 vs 3.53 ng/mg; P<.001) (Figure 2). Percent drug release of EES vs R-ZES was similar (P=.11). Circulating drug levels peaked at 30 minutes for both stent groups at similar levels, with steadily decreasing levels observed at 3 and 24 hours after deployment.

Morphometric analysis. Area measurements of external elastic lamina and underlying plaque were similar between EES and R-ZES  groups (Table 2). Stent area was significantly less in the EES group vs the R-ZES group at both 28-day follow-up (6.12 ± 0.32 mm² vs 7.09 ± 0.57 mm²; P=.004) and 60-day follow-up (5.08 ± 0.83 mm² vs 5.91 ± 0.61 mm²; P=.043). There were no significant differences in mean neointimal thickness above struts or neointimal area between both groups. Percent area stenosis increased at similar amounts from 28 days to 60 days in both groups (Table 2). Overall, both stent groups showed similar morphometric characteristics.

Histologic analysis. At 28 days, EES and R-ZES groups had similar rates of uncovered struts (21.66 ± 9.5% vs 24.91 ± 7.55%; P=.43); however, by 60 days, the number of uncovered struts decreased in the EES group, whereas in the R-ZES group, the incidence remained similar (15.83 ± 11.99% vs 25.67 ± 19.97%; P=.22). Both groups showed mild-to-moderate fibrin deposition without a significant difference between both groups. The EES group showed lower number of inflammatory cells adherent to the luminal surface at both 28 (P=.30) and 60 days (P=.10). Significantly lower inflammation score (P=.044) and giant cell infiltration (P=.014) were observed in the EES group vs the R-ZES group at 28 days. However, by 60 days, the inflammation and giant cell reaction were similar between groups (Table 3). The area of RAM11-positive macrophages in the neointima was significantly less in the EES group vs the R-ZES group at 28 days (P=.014); areas were similar between the two groups at 60 days (P=.73) (Figure 3). 

Endothelial coverage by scanning electron microscopy. At 14 days, reendothelialization above struts was similar between the EES (38 ± 39%) and R-ZES (40 ± 28%) groups. Stent struts lacking endothelial coverage generally showed focal aggregates of platelets and inflammatory cells, including giant cells. To a limited degree, adherent inflammatory cells and platelets were also found on endothelial surface between stent struts, but no differences were noted between the two groups (Figure 4). Areas between struts showed >88% endothelial coverage in both groups. At 28 days, reendothelialization above struts increased in both EES and R-ZES groups to similar percentages (79 ± 24% vs 81 ±14%; P=.81) with almost full coverage between struts >90%.

Endothelial maturation by confocal analysis. Areas of endothelial cells with PECAM-1 localized to cell-to-cell contact sites above struts were similar between EES and R-ZES groups at 14 days (11 ± 11% vs 5 ± 5%; P=.39). Percentages of cells expressing PECAM-1 between struts were similar, although the values for EES were greater (~18% vs ~7%; P=.25), but remained insignificant. At 28 days, PECAM-1 expression above struts was significantly higher in the EES group vs the R-ZES group (33 ± 20% vs 22 ± 21%; P=.040) (Figure 5). Between struts, a trend toward higher PECAM-1 expression was observed in the EES group vs the R-ZES group (34 ± 24% vs 20 ± 21%; P=.057). TM expression was observed at 28 days in the areas where PECAM-1 was positive. TM was only weakly expressed in both groups; therefore, the quantitative or qualitative analyses were not performed.

Discussion

A head-to-head comparison of the second-generation EES and R-ZES groups using the healthy and atherosclerotic rabbit ilio-femoral artery model was conducted. This comparison showed equivalency in terms of neointimal growth between arteries implanted with the two stents. However, endothelialization rates differed between the two stents since greater endothelial cell maturation was observed in EES vs R-ZES implanted arteries, even though strut coverage was similar. Furthermore, significantly less inflammation was observed in the EES vs R-ZES groups, which parallels the endothelial maturation findings at 28 days. These results highlight the first preclinical findings of the second-generation everolimus- and slow-release zotarolimus-eluting stents. 

 Preclinical testing in animal models is an important part of the regulatory process and is used to determine the safety and projected efficacy of devices.⁶ Additionally, even after commercialization of a device, a comparative understanding may shed light on observations seen clinically. Comparative preclinical histological studies remain the most effective method for assessing the vascular responses to various devices. Although non-diseased models are well accepted in assessing device performance, atherosclerotic models are preferred to distinguish differences in vascular inflammation and healing between comparator DESs.⁷ We therefore utilized the atherosclerotic rabbit ilio-femoral artery model to discern any potential differences between the second-generation EES and R-ZES groups with respect to inflammation and vascular healing and used a non-diseased model for assessing reendothelialization, as well as drug release and retention.

Morphometric comparison of the two stent groups showed similar neointimal growth and percent area stenosis at 28 and 60 days using the atherosclerotic model. These results are in agreement with recent clinical findings, in which EES and R-ZES groups were compared head-to-head in a randomized controlled trial in real-world patients (target lesion revascularization at 1 year: 8.3% for EES vs 8.2% for R-ZES; P=<.001 for non-inferiority).¹³ However in our preclinical study, despite similarities in neointimal growth, the inflammatory response differed between the stents, since significantly greater peristrut inflammation and foreign-body giant cells were observed in the R-ZES vs EES groups at 28 days. Increased inflammatory reaction of DESs has been shown to be related to polymer coating in first-generation DESs, but not in second-generation devices.¹³ In contrast to the well-known history (>20 years) of the EES fluoropolymer in biomedical applications, the BioLinx polymer of the R-ZES remains relatively unknown. However, in this model, inflammation rates were similarly resolved in both groups at 60 days. Responses to polymer coatings are complex and without controlled prospective evaluation of individual components (in particular, polymer coating without the drug) are difficult to isolate any adverse reaction. Furthermore, given that both devices have similar drugs, the higher drug dose in the R-ZES group versus the EES group could further prolong the inflammatory response at later time points. 

Macrophage content was observed to be greater in the plaque as compared to the neointima in both stent groups, which supports previous published work that mTOR inhibition by everolimus selectively clears macrophages but not smooth muscle cells through autophagy.¹⁴ The clearance of macrophages within the plaque underlying the stent (neointima) but not in remote regions (media) is consistent with our results. The lack of secondary necrosis or inflammation suggests an autophagic mechanism of macrophage clearance. This preferential mechanism leads to selective, clean removal of macrophages within the plaque, potentially minimizing contribution to necrotic core formation, inflammation, and thrombosis. Within the neointima, there was a significantly greater amount of macrophage in the R-ZES group vs the EES group at 28 days; however, the total content was low for both groups. 

Morphometric measurements also demonstrated a significantly smaller stent area in the EES group vs the R-ZES group at both 28 and 60 days after implantation. Since both stents were similar in size (diameter, 3.0 mm; length, 12 mm) and deployment parameters (nominal pressure for 30 seconds), these differences could be related to stent design and stiffness. Both the EES and R-ZES use cobalt-based alloys for their stent platform (the L-605 and F-562 alloy, respectively). The R-ZES has a slightly greater strut thickness (91 µm vs 81 µm in the EES) which may contribute to a more stiff/rigid stent.¹⁵ 

Clinically, non-endothelialized stent struts have been shown to be a major risk factor for late stent thrombosis.¹⁶ The assessment of reendothelialization rates has therefore become a major part of device evaluation in preclinical studies. Moreover, the EES has repeatedly had lower stent thrombosis rates than the R-ZES out to 1 year,¹³ and comparative studies in preclinical models may shed light on observations seen clinically. The rabbit iliac artery model is the preferred model, since the iliac arteries endothelialize more slowly compared to pig coronaries, and it is therefore a better model to assess potential differences between devices.⁶ In comparing the EES and R-ZES, SEM analysis demonstrated similar endothelial coverage between and above struts at 14 and 28 days. The expression of CD31, however, was shown to be significantly greater for the EES vs the ZES at 28 days. Differences in the expression of CD31 suggest a more rapid gain of endothelial maturation, which has been observed previously for BMS and EES implanted arteries.⁹ 

The drug dose and release properties can affect endothelial behavior and maturation. The PK analysis showed similar release curves between the two groups; however, arterial drug concentration was approximately one order of magnitude higher in the R-ZES group. Although differences exist in loading drug amount (100 ug/cm² vs 160 ug/cm²), zotarolimus is more lipophilic than everolimus, which is the likely cause of the greater arterial drug retention.¹⁷ Differences in drug release and retention could explain the delay of endothelium maturation observed within the R-ZES group vs the EES group. 

Study limitations. Although the present study utilized both the normal and atherosclerotic diseased models, they are not entirely representative of lesions responsible for coronary artery disease in man. Moreover, the achieved levels of circulating cholesterol levels in rabbits are considerably higher, which do not accurately represent the lower values applicable in humans. Vascular healing, ie, endothelialization, occurs more rapidly in both the healthy and atherosclerotic rabbit models than in atherosclerotic human coronary arteries. There were numerical differences in some of the biological endpoints that did not reach statistical significance, presumably due to the limited number of animals used in this study. Therefore, further study in a larger cohort may be warranted to clearly determine the biological effects of both stents.

Conclusion

The present findings suggest similar neointimal progression between the second-generation EES and R-ZES drug-eluting stents, resembling results from clinical studies. Histopathologic evaluation at 28 days showed greater inflammation and lower endothelial maturation in the R-ZES group as compared to the  EES group. These observations could be related to the increase arterial drug deposition observed in the R-ZES group vs the  EES group. 

References

1. Morice MC, Serruys PW, Sousa JE, et al. A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization. N Engl J Med. 2002;346(23):1773-1780.
2. Joner M, Finn AV, Farb A, et al. Pathology of drug-eluting stents in humans: delayed healing and late thrombotic risk. J Am Coll Cardiol. 2006;48(1):193-202.
3. Farb A, Sangiorgi G, Carter AJ, et al. Pathology of acute and chronic coronary stenting in humans. Circulation. 1999;99(1):44-52.
4. Nakazawa G, Ladich E, Finn AV, Virmani R. Pathophysiology of vascular healing and stent mediated arterial injury. EuroIntervention. 2008;4(Suppl C):C7-C10.
5. Schwartz RS, Chronos NA, Virmani R. Preclinical restenosis models and drug-eluting stents: still important, still much to learn. J Am Coll Cardiol. 2004;44(7):1373-1385.
6. Nakazawa G, Finn AV, Ladich E, et al. Drug-eluting stent safety: findings from preclinical studies. Expert Rev Cardiovasc Ther. 2008;6(10):1379-1391.
7. Nakazawa G, Nakano M, Otsuka F, et al. Evaluation of polymer-based comparator drug-eluting stents using a rabbit model of iliac artery atherosclerosis. Circ Cardiovasc Interv. 2011;4(1):38-46.
8. Ribichini F, Joner M, Ferrero V, et al. Effects of oral prednisone after stenting in a rabbit model of established atherosclerosis. J Am Coll Cardiol. 2007;50(2):176-185.
9. Joner M, Nakazawa G, Finn AV, et al. Endothelial cell recovery between comparator polymer-based drug-eluting stents. J Am Coll Cardiol. 2008;52(5):333-342.
10. Schwartz RS, Edelman ER, Carter A, et al; Consensus Committee. Drug-eluting stents in preclinical studies: recommended evaluation from a consensus group. Circulation. 2002;106(14):1867-1873.
11. Scheller B, Speck U, Romeike B, et al. Contrast media as carriers for local drug delivery. Successful inhibition of neointimal proliferation in the porcine coronary stent model. Eur Heart J. 2003;24(15):1462-1467.
12. Bailer AJ. Testing for the equality of area under the curves when using destructive measurement techniques. J Pharmacokinet Biopharm. 1988;16(3):303-309.
13. Serruys PW, Silber S, Garg S, et al. Comparison of zotarolimus-eluting and everolimus-eluting coronary stents. N Engl J Med. 2010;363(2):136-146.
14. Verheye S, Martinet W, Kockx MM, et al. Selective clearance of macrophages in atherosclerotic plaques by autophagy. J Am Coll Cardiol. 2007;49(6):706-715.
15. Zahedmanesh H, Lally C. Determination of the influence of stent strut thickness using the finite element method: implications for vascular injury and in-stent restenosis. Med Biol Eng Comput. 2009;47(4):385-393.
16. Finn AV, Joner M, Nakazawa G, et al. Pathological correlates of late drug-eluting stent thrombosis: strut coverage as a marker of endothelialization. Circulation. 2007;115(18):2435-2441.
17. Burke SE, Kuntz RE, Schwartz LB. Zotarolimus (abt-578) eluting stents. Adv Drug Deliv Rev. 2006;58(3):437-446.