Biomaterials-Enabled Antithrombotics: Recent Advances and Emerging Strategies

Abstract

Antithrombotic and thrombolytic therapies are used to prevent, treat, and remove blood clots in various clinical settings, from emergent to prophylactic. While ubiquitous in their healthcare application, short half-lives, off-target effects, overdosing complications, and patient compliance continue to be major liabilities to the utility of these agents. Biomaterials-enabled strategies have the potential to comprehensively address these limitations by creating technologies that are more precise, durable, and safe in their antithrombotic action. In this review, we discuss the state of the art in anticoagulant and thrombolytic biomaterials, covering the nano to macro length scales. We emphasize current methods of formulation, discuss how material properties affect controlled release kinetics, and summarize modern mechanisms of clot-specific drug targeting. The preclinical efficacy of these technologies in an array of cardiovascular applications, including stroke, pulmonary embolism, myocardial infarction, and blood contacting devices, is summarized and performance contrasted. While significant advances have already been made, ongoing development efforts look to deliver bioresponsive “smart” biomaterials that will open new precision medicine opportunities in cardiology.

Introduction

Blood clotting dysfunctions, and the complications that result, are a tremendous burden on the healthcare system. They account for up to $10 billion in costs annually, and per event treatment costs can be as high as $20,000.¹ The human costs, however, are far greater. One person dies every 5 min in the United States due to blood clotting complications, accounting for more annual deaths than AIDS, breast cancer, and automobile accidents combined.²

Coagulation occurs through two routes, intrinsic and extrinsic, which converge to a common pathway.³ The extrinsic pathway is activated during tissue damage and so is the primary mode of clot initiation. Following damage, a cascade of proteases and glycoproteins catalyze reactions that ultimately cleave fibrinogen to produce fibrin. Hierarchical assembly of fibrin monomers into a supramolecular mesh leads to the stabilization of platelet aggregates and the formation of an anchored clot. The resulting thrombus impedes blood flow and tissue oxygenation, causing ischemia and stroke.

Clinically, blood clots are managed via two complementary strategies: prevent and treat.⁴⁻⁷ Antiplatelets and anticoagulants are preventative agents that inhibit blood clot formation through direct and indirect mechanisms.⁸ Antiplatelets are direct pharmaceuticals that act by inhibiting platelet aggregation to prevent emboli formation.⁹ This therapeutic class includes various cyclooxygenase inhibitors (e.g., aspirin), adenosine diphosphate receptor agonists, and glycoprotein IIB/IIA inhibitors, among others. Anticoagulants, conversely, inhibit specific factors within the coagulation cascade to indirectly modulate clotting.¹⁰ Vitamin K antagonists (e.g., warfarin) and the glycosaminoglycan heparin were the earliest to be approved and remain the mainstay of clinical use today. Warfarin acts by broadly inhibiting Factors II, VII, IX, and X, while heparin inactivates thrombin and Factor Xa. Although effective, the promiscuity of these agents increases the risk of adverse bleeding events and hemorrhage in compromised patients. As a result, the last two decades have seen the development of several targeted oral anticoagulants that precisely inhibit specific clotting factors (Figure 1).¹⁰⁻¹⁴

Figure 1.

Selected landmarks in anticoagulant development over the last eight decades. Year of drug approval or of completion of phase 2 trials is shown. Figure reproduced with permission from ref (10). Copyright 2021 International Society on Thrombosis and Hemostasis.

While anticoagulants are used to prevent blood clotting, thrombolytics are drugs that rapidly dissolve an established clot. Thrombolytics are serine proteases that cleave plasminogen to form the fibrinolytic enzyme plasmin. Tissue plasminogen activator (tPA), streptokinase, and urokinase are the most well-known examples and are commonly employed as emergency treatments for stroke, heart attack, and pulmonary embolism.¹⁵

The pharmacologic limitations of anticoagulants and thrombolytics, which include rapid renal clearance, short half-lives, and off-target toxicity, as well as clinical barriers of dosing errors and patient compliance, have made these compounds attractive candidates for biomaterials-enabled controlled delivery applications. In particular, both oral warfarin and intravenous heparin must be carefully dosed for the patient’s body weight and clotting sensitivity.¹⁶⁻¹⁸ Patient variability, user error, and infuser malfunctions can all lead to anticoagulant overdosing. Additionally, the short plasma half-life of these agents requires frequent maintenance dosing, further increasing the risk of dosing errors. Major adverse events include severe thrombocytopenia, hemorrhage, and death, particularly in small children, cancer patients, and pregnant women.¹⁹⁻²¹ Biomaterial scaffolds that encapsulate and controllably release anticoagulants and thrombolytics offer the potential to reduce dosing frequency, improve patient compliance, minimize the need for postadministration monitoring, extend therapeutic plasma concentrations, and reduce off-target distribution and adverse effects (Figure 2).

Figure 2.

Various nano- through macroscale biomaterials have been engineered to deliver antithrombotic agents to vascular embolisms, with the goal of controlling spatiotemporal drug localization and release kinetics. The two most widely employed strategies, nanoparticle delivery (blue) and implantable patches/gels (gray), are depicted. For nanoparticle platforms, the antithrombotic agent (orange) is loaded within the carrier interior, onto the surface of the particle, or as a building block of the vehicle itself. Patches and hydrogels rely on stimuli-responsive diffusion of the therapeutic agent across dermal and endothelial interfaces to achieve systemic delivery.

This review will survey recent developments in biomaterials design across nano, micro, and macro length scales in the context of antithrombotic cardiovascular therapies. The formulation of nano- and microparticles, hydrogels, patches, and several other systems used to deliver anticoagulants and thrombolytics will be discussed, with merits and drawbacks highlighted. Various strategies employed to preferentially localize these biomaterial-supported therapies to sites of thrombosis, thereby reducing drug-associated adverse events, will be presented. Finally, the in vitro, ex vivo, and in vivo performance of the materials will be summarized and compared to current clinical practice.

It should be noted that, in addition to their cardiovascular applications, several anticoagulant-functionalized biomaterials, particularly incorporating heparin, have been developed as tissue engineering scaffolds, therapeutics in oncology and acute lung injury, and as anti-inflammatory agents. These nonthrombotic applications have been comprehensively reviewed elsewhere,²²⁻²⁵ and so this review will be limited to technologies with explicit anticoagulant and thrombolytic functions.

Nanoscale Vehicles

Nanotechnology is a cornerstone of modern drug delivery approaches, as nanoscale carriers can improve drug solubility and modulate pharmacokinetic properties, possess high carrying capacities due to their large surface area to volume ratios, and enable preferential transport of therapeutic cargo to specific sites.²⁶⁻²⁸ In the context of thrombosis, nanoparticles show unique effects on coagulation factors, platelets, endothelial cells, and leukocytes.^29,30 If engineered appropriately, these properties can potentiate beneficial interactions between nanoparticles and constituents of the coagulation system to intervene in clotting disorders. Further, nanoparticles are the only delivery vehicle that can diffuse into the fibrin network of developing clots, which possess pore sizes that range from 1 nm to 1 μm,^31,32 to affect their bioactive functions within the thrombi interstitium.

In addition to their size-dependent effects, the material used to construct nanoparticle delivery vehicles can impart unique benefits to the system. Polymers and lipids remain favored building blocks due to their scalability, ease of chemical functionalization, and safety profiles. However, magnetic materials have recently gained increasing attention based on their ability to preferentially localize thrombolytic and anticoagulant cargo to thrombi under the guidance of a magnetic field. Finally, biological constituents (e.g., proteins and peptides) are emerging as next generation materials, as they can be engineered with bioactive properties that productively complement the function of loaded thrombolytics and anticoagulants. In this section, we review recent efforts to utilize each of these building blocks in the construction of antithrombotic nanotechnologies and summarize the resulting material–function–performance relationships.

Polymers

Poly(lactide-co-glycolide) (PLGA) nanoparticles have been one of the most widely employed platforms for vascular applications due to their chemical versatility, biodegradability, and toxicologic safety of the degradation products.³³⁻³⁵ For example, Lee et al. developed a layer-by-layer PLGA particle platform incorporating heparin and glutathione (GSH) to elicit anticoagulant and antioxidant therapy following ischemia/reperfusion injury.³⁶ Here, the PLGA nanoparticle core was functionalized with stearylamine to provide a positive surface charge that enabled electrostatic complexation of anionic heparin. This was then followed by adsorption of the cationic conjugate formed via ligation of GSH to polyethylenimine (PEI), and finally, the particle was coated with the negatively charged carbohydrate hyaluronic acid (HA). This final surface modification was done to enable targeted delivery of the particle to HA-receptors expressed on the surfaces of mesenchymal stem cells at the injury site. The resulting particles, which were approximately 100–150 nm in size, showed sustained release of heparin with first order kinetics, achieving ∼10% total release in physiologic media after 96 h. This is a significant advance given the short plasma half-life of heparin (t_1/2 = ∼1 h), requiring maintenance dosing every 1 to 4 h in the clinical setting. Additionally, clotting inhibition matched the activity of an equivalent concentration of free heparin, indicating that the vehicle did not impair the bioactivity of the anticoagulant.

Acrylamide polymers are another popular platform to develop anticoagulant nanomaterials. Xu et al. designed a thrombin-responsive acrylamide derivative that enabled thrombi-regulated delivery of a recombinant hirudin anticoagulant (Figure 3).³⁷ To develop the thrombin-sensitive polymer scaffold, acrylamide monomers were polymerized in the presence of a clot-targeting peptide (allyl-GGCR(NMe)EKA) and thrombin cleavable peptide (TCP) cross-linker (allyl-GGGLVPRGSGGG-allyl). Recombinant hirudin was subsequently encapsulated within the ∼50 nm nanoparticles following acrylamide-peptide assembly. In vitro experiments confirmed selective hirudin release following thrombin-mediated cleavage of TCP linkers, without appreciable release from control formulations prepared using a noncleavable linker analogue. Additional clot binding assays showed that display of the clot-targeting CR(NMe)EKA peptide from the surface of the nanoparticle increased intraclot accumulation by nearly 6 times relative to nontargeted control particles. These bioactive properties converged to improve survival in a mouse pulmonary embolism model from 30% for saline treated controls to 100% for the particle formulation.

Figure 3.

Depiction of the construction and antithrombotic activity of hirudin-loaded polymeric nanoparticles. (A) Recombinant hirudin variant 3 (HV) is encapsulated within clot-targeted thrombin-responsive nanogels (HV/ctNGs) during cross-linking of acrylamide monomers (Aam) and peptides. (B) Thrombin-responsive delivery is achieved through a closed-loop strategy in which thrombin activation leads to HV release from the carrier. The liberated anticoagulant subsequently inhibits thrombin to evoke a negative feedback loop that suppresses drug release from the bioresponsive carrier. Figure reproduced with permission from ref (37). Copyright 2020 Xiao Xu et al. Figure is licensed under CC BY-NC 4.0, no changes were made to the graphic.

Carbohydrate polymers have also been employed to develop electrostatic nanocomplexes that modulate the stability, bioavailability, and pharmacokinetics of anticoagulants. Liang and co-workers, for example, complexed cationic chitosan, a mucoadhesive polymer, with the pH-sensitive anionic polymer hydroxypropyl methylcellulose phthalate (HPMCP) to create gastric stable polyplexes.³⁸ Heparin present in the mixture was directly encapsulated within the particles during ionic assembly. This approach leverages the insolubility of HPMCP at pH < 5.0 to maintain stable polyplexes within the acidic gastric environment. As a result, heparin is retained within the particle to prevent hydrolytic and enzymatic degradation of the therapeutic within gastric acid, which otherwise is a barrier that prevents its oral administration. Once transported to the neutral intestinal compartments, the HPMCP becomes charged and solubilized by the gastrointestinal fluid to release the encapsulated heparin and chitosan constituents. Adhesion of the released chitosan to the mucosal lining opens epithelial tight junctions in the gastrointestinal tract to enhance the systemic permeation and bioavailability of loaded heparin. Biophysical and biochemical assays demonstrated that the 300–500 nm diameter particles had a heparin encapsulation efficiency of >90% and were able to reduce the diffusion and degradation rate of heparin in simulated gastric fluid by 2-fold, relative to the free therapeutic. In vivo studies using Sprague–Dawley rats showed that the mucoadhesive nanoparticles extended anticoagulant effects relative to free heparin after a single dose (50 mg/kg equivalent heparin) administered via oral gavage. As an example, activated partial thromboplastin assays performed 8 h after administration demonstrated a 2.5-fold increase in blood clotting time for animals orally dosed with the chitosan-HPMCP-heparin nanoparticle over free heparin.

In a related study, cationic chitosan was again used to form anticoagulant polyplexes, this time using Enoxaparin (Enox) as the anionic counterpart.³⁹ Like heparin, Enox is a highly sulfonated polysaccharide that activates antithrombin to prevent coagulation, but with the benefit of an extended circulatory half-life. Formulation optimization studies revealed that mass ratios of 1:2.5 Enox:chitosan produced 156 nm complexes, which could be further colloidally stabilized and optimized upon addition of free dextran sulfate. Importantly, this method yielded nanocomplexes <200 nm in size, which the authors argued would reduce particle clearance by the mononuclear phagocytic system. Subsequent in vitro characterization confirmed ionic complexation of Enox achieved sustained anticoagulant release, with <35% liberated after a 14-day incubation. Follow up in vivo studies measuring Factor Xa activity showed that chitosan polyplexes extended the anticoagulant activity of Enox from 5 to 135 h, which was further increased to >144 h using particles coformulated with dextran sulfate. Similar results were observed for nanoparticles prepared from mixing dextran sulfate and the chemotherapeutic Doxorubicin, where the particulate formulation significantly reduced clot rate and thrombus size in in vitro and in vivo models.⁴⁰ In sum, these studies demonstrate the potential for polymeric nanodelivery systems to stabilize loaded anticoagulants, extend their half-life in serum and other physiologic fluids, and open new oral administration routes for what would otherwise be parenteral restricted therapies.

Lipids

Liposomal carriers have enjoyed widespread use in drug delivery, with renewed attention gained due to their recent use in COVID-19 mRNA vaccines.⁴¹⁻⁴³ These applications often leverage the ease of formulation, safety, and cargo loading versatility of lipid nanoparticles. Although liposome size can be readily varied from 50 nm to 5 μm, particles used for cardiovascular applications are commonly 100–500 nm in size. Endreas et al. showed that a fatty-acid-functionalized peptidomimetic anticoagulant could be readily appended to the surface of ∼350 nm liposomes formulated with DPPC (1,2-dipalmitoyl-sn-glycero-3-phosphocholine), CH (cholesterol), and the clotting inhibitor at a ratio of 65:30:5 mol %.⁴⁴ Plasma clotting times confirmed that unloaded liposomal formulations were inactive, while those loaded with the inhibitor possessed nanomolar anticoagulant potency. Additional formulation screening revealed that incorporating a flexible polyethylene glycol (PEG) spacer between the inhibitor and the liposomal carrier improved its anticoagulant activity by 100-fold. In a separate study, the anticoagulant transmembrane protein thrombomodulin was inserted into liposomal membranes to prepare a biomimetic and nanoscale analogue of endothelial cells.⁴⁵ Thrombomodulin is an anticoagulant protein expressed at the surfaces of endothelial cells to alter the function of thrombin protease from a procoagulant to an anticoagulant state. This is an essential feedback mechanism during vascular repair that prevents excessive coagulation. By reconstituting a recombinant form of thrombomodulin into liposomes, the authors showed that they could extend the proteolytic stability of the protein from 3 h to >50 h. Moreover, incorporation of PEG into the liposomal formulation was again found to preserve the full function of the protein relative to non-PEGylated liposomal analogues, which experience a 25% reduction in anticoagulant effects relative to the native protein. However, this increased activity came at a cost, as PEGylating the liposomal carrier also led to more rapid degradation of the loaded thrombomodulin anticoagulant by plasma proteases. These studies exemplify the importance of optimizing the conformational flexibility of anticoagulants loaded into liposomes, necessitating empirical optimization of nanoparticle surface sterics to balance the stability and bioactivity of the therapeutic.

An interesting alternative is the development of bioresponsive liposomes that react to the thrombus environment to deliver therapeutics on-demand. Huang and co-workers demonstrated the potential of this approach using nanoliposomes loaded with the thrombolytic tPA and functionalized with cyclic arginine–glycine–aspartic acid (cRGD) peptides.⁴⁶ Here, cRGD was included as a targeting ligand to enable preferential particle binding to α_IIbβ₃ integrins overexpressed on the surfaces of platelets activated during thrombosis. Flow cytometry and fluorescence resonance energy transfer (FRET) experiments demonstrated that not only did the display of cRGD enable preferential liposomal binding to activated platelets but also cRGD–integrin interactions triggered fusion of the lipid carrier with the platelet membrane. This behavior allowed for the programmable delivery of tPA to activated platelets at the site of thrombosis. As a result, tPA-loaded cRGD liposomes showed complete lysis of clots in vitro after 1.5 h of incubation, while non-cRGD controls achieved only 35% lysis at the same time point. Not surprisingly, inclusion of PEG further enhanced the fibrinolytic activity of the liposomal platform, decreasing the time required for dissolving 50% of the clot from 122 min for the native tPA-loaded liposomes to 48 min for formulations modified with the amphiphilic polymer.

Magnetite

Magnetite (Fe₃O₄) is a ferrimagnetic iron oxide that possesses several unique properties advantageous to cardiovascular applications.⁴⁷⁻⁴⁹ First, magnetite nanoparticles can be synthesized with diameters ≤100 nm, which allows them to readily diffuse throughout the thrombi fibrin matrix. The superparamagnetic properties of magnetite nanoparticles can also be exploited for imaging purposes and to magnetically direct particle localization in tissues. Finally, under a magnetic field, magnetite can be controllably heated to enable thermal mediated drug delivery. Zhao et al. exploited these properties to prepare heparin-coated Fe₃O₄ nanoparticles via a layer-by-layer approach that utilized cationic PEI as a binding mediator.⁵⁰ Hemodynamic studies show that heparin coatings dramatically decreased the hemolytic activity of the 80 nm magnetite nanoparticles and produced sustained anticoagulant effects in canine models. Other work by Prilepskii and co-workers utilized heparin-coated magnetite nanoparticles (∼100 nm) to adsorb the thrombolytic urokinase-type plasminogen activator (uPA) to the particle surface for combinatorial anticlotting activity.⁵¹ These interactions were mediated via binding of the uPA kringle domain with the heparin surface coating, thereby loading the thrombolytic onto the particle carrier without relying on permanent chemical conjugation or weak electrostatic interactions. Next, the authors tested the thrombolytic potential of uPA-loaded particles in a peristaltic flow loop. Here, an external magnet was placed adjacent to the in-loop formed clot and optical microscopy used to monitor clot size as a function of time. Results showed a significant enhancement of clot lysis when uPA was delivered via magnetite particles, reaching 100% clot removal after 35 min. Approximately 60% of the clot was dissolved for control samples treated with free uPA at the same time point. This is significant, as it suggests magnetite-based nanothrombolytics could be useful in emergency treatments of stroke, heart attack, and pulmonary embolisms, where the rapid clot removal achieved by this platform could dramatically improve patient outcomes.

Interestingly, the authors found only a moderate change in clot dissolution rates when the particles were exposed to the external magnetic field relative to control samples left unmagnetized. This suggests that biochemical binding of heparin and/or uPA to the clot is a stronger localization mechanism relative to the physical magnetic effects. As a result, the authors conducted subsequent in vivo thromboembolism experiments in rats and rabbits utilizing uPA-loaded Fe₃O₄ nanoparticles without additional magnetic localization. Results show that uPA-loaded particles restored perfusion four times faster than free uPA in both a rat carotid artery and rabbit femoral artery embolism model. Biodistribution and toxicity analyses in these animals showed that the particles were primarily cleared by the liver without sustained toxicity up to 30 days post treatment, even at particle concentrations 3 orders of magnitude greater than the therapeutic dose.

Proteins and Peptides

Biomacromolecules are emergent building blocks in the design of antithrombotic nanomaterials, as they can be engineered to yield nonspherical hierarchical nanostructures that exert shape-dependent vascular targeting.^52,53 For example, rod-shaped tobacco mosaic virus (TMV) was five times more likely to bind to carotid artery emboli in a thrombosis mouse model compared to sphere-like cowpea mosaic virus particles (CPMVs).⁵⁴ Based on these findings, TMV nanocarriers, ∼200 nm in their longest axis, were conjugated to tPA using a bifunctional PEG linker and tested for thrombolytic activity.⁵⁵In vitro and in vivo studies confirmed that TMV-loaded tPA retained its fibrinolytic activity but showed four times greater accumulation on the surface of developing emboli relative to free tPA. Additional experiments using a murine arterial thrombosis model demonstrated that animals treated with TMV-tPA particles showed more frequent and rapid reopening of occluded vessels compared to saline treated controls. Importantly, the improved precision of the platform also led to reduced bleeding complications due to tPA therapy, with an average bleeding time of 430 s for TMV-tPA particles and 858 s for free tPA. Similar results were found from analogous studies on TMV-delivered streptokinase.⁵⁶

An additional advantage of biologic platforms is that they can be engineered with intrinsic bioactivities that augment or enhance the therapeutic function of anticoagulant and thrombolytic cargo. For example, our group has shown that peptide-based nanoemulsions displaying RGD, referred to as nanopeptisomes, preferentially accumulate at the surfaces of growing thrombi and have intrinsic anticoagulant functions under ultrasound (Figure 4).⁵⁷ Here, competitive saturation of platelet surface sites by binding of particles to α_IIbβ₃ integrin receptors inhibited the further aggregation of activated platelets, thereby exerting antithrombotic functions even in the absence of a delivered anticoagulant or thrombolytic. Additional mechanical effects induced by particles oscillating under ultrasound enabled simultaneous clot dissolution. Peptides have also been used as a binding agent to mediate transport of anticoagulants via nature’s own cardiovascular delivery vehicle, red blood cells (RBCs). In a recent example, thiolated polylysine (PLL) was mixed with heparin to prepare nanoparticles stabilized via spontaneous disulfide (S-S) cross-linking.⁵⁸ This strategy is intended to slow heparin release via disulfide hydrolysis to maintain therapeutic concentrations in circulation. The nanocomplexes were then electrostatically adsorbed onto the surface of negatively charged RBCs to take advantage of the long-circulation properties of erythrocytes. Optimization experiments demonstrated RBC attachment was successful when the weight ratio of PLL-to-heparin is ≥1.2; otherwise, the particles are negatively charged and do not electrostatically complex to anionic erythrocytes. Heparin release kinetics were then studied for PLL–heparin formulations prepared without cross-links (electrostatics only), disulfide cross-linked formulations, and particles adsorbed to the surface of RBCs. Disulfide cross-linking was found to prolong complete release of the anticoagulant from 24 h for non-cross-linked controls to 96 h for the cross-linked formulations. By binding to RBCs, heparin release was further slowed to only 80% delivery achieved over 120 h of incubation. Complexation of cross-linked PLL–heparin nanoparticles to RBCs imparted the system with several other advantages. For instance, the micron size of the RBC–particle complex limited extravascular diffusion of the nanoparticles during circulation, thereby minimizing their off target biodistribution. RBCs also are evolutionarily optimized to avoid clearance by macrophages, which the authors demonstrated by showing a nearly doubled serum persistence time of heparin when complexed to RBCs compared to free nanoparticles or heparin alone. In addition to its clever use of micron-sized RBCs as the delivery agent, this work illustrates the unique benefits of micron-scale carriers over their nanoparticle counterparts, a design aspect we further discuss in the following section.

Figure 4.

Ultrasound-responsive nanopeptisomes for synchronous clot detection and antithrombosis in deep vein thrombosis. Peptide nanoemulsions are formulated via assembly of a fibrinogen-mimetic fluoropeptide emulsifier (blue) at the interface of perfluorocarbon nanodroplets (green). In circulation, nanopeptisomes preferentially bind to the surfaces of activated platelets to enable spatially resolved Doppler US imaging and simultaneous competitive inhibition of clot growth. Reproduced with permission from ref (57). Copyright 2021 Wiley-VCH GmbH.

Microscale Vehicles

Unlike nanoscale carriers, microscale vehicles are largely prevented from passing through endothelial tight junctions and are slower to be phagocytosed by circulating macrophages. As a result, microparticles are less likely to be cleared from the vasculature relative to nanoparticles and, therefore, circulate longer and have a greater probability of interacting with clots. These advantages are further complemented by a high particle volume and more facile formulation procedures, often leading to greater loading capacities of antithrombotics. Salama et al. exploited these properties to develop a polymer-based microcarrier encapsulating the antiplatelet agent dipyridamole (DIP).⁵⁹ The designed system relies on free radical polymerization using acrylamide and a variety of polymethacrylate-based copolymers to form a polymer network, with DIP sequestered within the mesh architecture of the carrier. Multifactorial design was used to optimize the acrylamide and cross-linker ratios, leading to a lead formulation that showed a DIP encapsulation efficiency of 96%. This system allowed for controlled release of DIP over an 8 h period under physiologic conditions, with a burst release of 34% at 1 h and 93% total release at 8 h. The lead microparticle formulation was then mixed with pectin and gellan gum to create a paste-like “raft” that remained buoyant in gastric fluids. As a result, after oral administration the particles were retained within the stomach where, at the acidic gastric pH, they controllably delivered DIP and allowed its diffusion through the gastric wall and into circulation. Pharmacokinetic studies in human volunteers showed that the microparticle DIP delivery platform increased the drug’s maximum plasma concentration by 1.5 times when compared to the commercial DIP product Persantin. Furthermore, there was a 2.3-fold increase in bioavailability of the drug due to microparticle delivery, ultimately allowing for a reduction in dosing frequency and improvement of patient compliance.

Yet, despite their potential to enable new oral formulations, microparticle anticoagulants have been limited in their use for intravascular applications. This is primarily due to the safety risk possessed by microparticles, which, if not sufficiently small and homogeneous in dimension, can function as microemboli that lodge in small capillary networks. Consequently, microparticle platforms have been used more frequently for procoagulant applications.⁶⁰⁻⁶⁴

Macroscale Materials

Delivery technologies that are millimeter in scale, or larger, offer the opportunity to locally deliver anticoagulants and/or thrombolytics to the vasculature from a static depot. We refer to these materials as “macroscale” vehicles, and their development has primarily focused on patch or hydrogel systems that can be subcutaneously injected, implanted, and/or adhered to the dermal layer. It is worth noting that these macrosystems have several advantages compared to their nano- and microparticle counterparts. These include facile preparation, relatively low cost, and ability to rapidly screen multiple formulations in a high throughput manner. Conversely, these implantable materials suffer from an inability to preferentially localize the delivered antithrombotics to the clot site; a property uniquely consigned to nanoparticle carriers.

Patches

Microneedle patches offer a less painful and minimally invasive alternative to standard intravenous administration. Transcutaneous delivery of anticoagulants and thrombolytics also provides sustained and controlled delivery of the cargo to circulation. This was recently demonstrated by Shen and co-workers, who designed a recombinant hirudin-loaded dissolvable polymer microneedle patch prepared from polyvinylpyrrolidone and poly(vinyl alcohol).⁶⁵ Hirudin was loaded into the polymer solution at a concentration of 25 μg/patch and then cast in a siloxane mold to create the microneedle features. The formed patch consisted of 225 (15 × 15) quadrangular pyramid shaped needles, each with a height of 589 μm. In vivo degradation assays in murine models confirmed that intradermal hydrolysis of the polymer microneedles led to delivery of 68% of the loaded hirudin over 20 min. The delivered hirudin was bioactive and showed an elongation of thromboplastin, prothrombin, and thrombin time by 1.4, 1.2, and 1.1 times, respectively, compared to the untreated control group. Importantly, the coagulation parameters for animals receiving the patch matched those treated with subcutaneous hirudin, indicating that the microneedle platform could serve as a more painless delivery system relative to traditional subcutaneous injection.

An additional advantage of microneedle platforms is their ability to enable bioresponsive approaches that deliver anticoagulants on demand. Zhang et al. recently reported a thrombin-responsive transcutaneous patch for programmable delivery of heparin (Figure 5A).⁶⁶ The design uses the thrombin cleavable peptide GGLVPRGSGGC as a cross-linker to ligate heparin (HP) and hyaluronic acid (HA). Entanglement of HP–peptide–HA chains forms a thrombin-responsive HA–HP (TR-HAHP) hydrogel network that was subsequently utilized to construct a microneedle patch (Figure 5B). The HP delivery feedback loop is initiated upon cleavage of the GGLVPR*GSGGC peptide by activated thrombin (star designates the thrombin cleavage site). The released HP then binds to and inactivates thrombin, which in turn negatively regulates peptide cleavage and closes the HP release feedback loop. Together, this approach allows release of heparin in the presence of activated thrombin, while suppressing its delivery in the absence of activated thrombin when it is not needed. Thrombin-mediated release was confirmed using fluorescently labeled heparin, and the results showed first-order release kinetics over 12 h (Figure 5C), with cumulative release scaling to the amount of thrombin present in solution. Parallel experiments demonstrated heparin release could be turned on/off via modulation of thrombin presence in solution (Figure 5D) and that heparin-loaded gels could sustainably release the anticoagulant over multiple hours (Figure 5E). The anticoagulant properties of TR-HAHP gels were next validated through a series of assays measuring intrinsic and extrinsic coagulation pathways, with results confirming similar anticoagulant functions of the gels to that of the free heparin control. With the functionality of the system confirmed, the HA–heparin mixture was embedded into a polymer matrix prepared from N,N′-methylenebis(acrylamide) and a photoinitiator in a silicone microneedle mold. The patch was prepared with 400 needles in a 12 × 12 mm² array. Thrombin-actuated release of heparin from the patch was then validated before investigating the performance of the technology in treating acute thrombosis in vivo. This work highlights the therapeutic potential of a feedback-controlled microneedle delivery system that dynamically responds to coagulation to provide drugs as needed, thereby potentially limiting the frequency of dosing errors compared to current delivery paradigms.

Figure 5.

(A) Design and mechanism of feedback-controlled heparin delivery from thrombin-responsive HA-HP conjugates (TR-HAHP). (B) Schematic of the TR-HAHP MN array patch in response to thrombin. (C) In vitro HP release from TR-HAHP hydrogels in the presence of 0–1.0 U/mL thrombin concentrations. (D) Pulsatile release of HP from TR-HAHP hydrogels in the presence (pink) and absence (blue) of thrombin. (E) Fluorescence microscopy images of a representative TR-HAHP hydrogel in thrombin solution, indicating loss of fluorescently labeled HP (green) at the indicated time points. Scale bar: 1 mm. Figure reproduced with permission from ref (66). Copyright 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

Hydrogels

Like microneedle technologies, several hydrogel platforms have been designed with cleavable networks to enable coagulation-responsive delivery of loaded antithrombotics. Maitz et al. conjugated heparin to star-shaped poly(ethylene glycol) (starPEG) through several cleavable peptides to prepare starPEG–heparin hydrogels.⁶⁷ Here, gels were prepared at a starPEG:heparin molar ratio of 1:1 and ligated via conjugation of the maleimide-functionalized starPEG with the C-terminal cysteine of the peptide. Heparin was then conjugated to the peptide’s N-terminus via carbodiimide chemistry to yield a cross-linked molecular network. Various peptides were included into the formulations to enable cleavage-associated heparin delivery via different activated coagulation factors, including FXIIa/Kallikrein, FXa, and thrombin. The performance of each formulation was then tested in the presence of equivalent concentrations of each respective coagulation factor. Hydrogels prepared from thrombin-responsive peptides showed the slowest release rate of 0.02 g cm^–2 s^–1 mol_enzyme^–1, while the FXa-responsive peptide had a 10 times faster release rate of 0.19 cm^–2 s^–1 mol_enzyme^–1. The kallikrein-responsive peptide hydrogel was 20 times more labile than the thrombin-responsive gel, with a release rate of 0.47 g · cm^–2 s^–1 mol_enzyme^–1. The slow-release rate of the thrombin-responsive hydrogel was attributed to the strong binding affinity of thrombin to heparin, which was based on observations that thrombin was predominately adsorbed to the hydrogel scaffold as opposed to degrading the matrix. Importantly, this work identified several thrombin-sensitive analogues that allowed control over thrombin-degradation kinetics, exemplified in a 3 h thrombin (Thr) exposure study showing that Thr_scrambled, Thr_slow, and Thr_fast had relative heparin release ratios of 0.2:1.0:2.7. Next, the team tested the anticoagulant activity of the different hydrogels upon incubation with whole blood. The FXIIa/Kallikrein-responsive hydrogel presented the lowest concentration of both F1+2, a marker for thrombin activity, and PF4, a marker of blood platelet activation. The Thr_fast- and FXa-sensitive hydrogels also exhibited anticoagulant activity but were not as effective as the kallikrein-responsive peptide hydrogels.

The same group went on to develop and characterize starPEG–heparin hydrogels prepared from thrombin-sensitive peptide cross-linkers.⁶⁸ Here they showed the importance of cross-linking density on hydrogel performance, with lower cross-linked hydrogels releasing more heparin and generating more potent anticoagulant responses relative to highly cross-linked networks. This behavior was attributed to the improved diffusion of thrombin within the low cross-linked networks. Interestingly, all formulations showed a steady, zeroth-order release of heparin from the gels in the presence of thrombin, with the release rate dependent on cross-link density. For example, cross-linked gels produced ∼100 μg/cm² of cumulative heparin release after 8 h, while highly cross-linked gels were defined by ∼10 μg/cm². Reducing this approach to practice, the team showed that coating polytetrafluoroethylene-based vascular grafts with the thrombin-responsive gel resulted in a significant reduction of pro-clotting activation relative to standard heparin-coated grafts.

Gritsch et al. further explored the importance of anticoagulant loading mechanism on hydrogel performance.⁶⁹ Using gelatin polymers cross-linked with acrylamide, the authors loaded heparin either by including it in the solution during cross-linking (chemical encapsulation) or adsorbing heparin after hydrogelation (physical encapsulation). In the latter method, the anionic heparin molecule electrostatically complexes with the cationic groups in gelatin to enable its loading into the gel. Degradation studies performed over 60 h showed that chemical encapsulation only led to substantive heparin release when the cross-linking stoichiometry of the gel was low (0.5). Conversely, physical loading produced hydrogels that controllably released the loaded polysaccharide over the entirety of the study period, with the rate and extent of release not significantly impacted by the material formulation. This work demonstrates how encapsulation processes can be altered to match the desired kinetics of anticoagulant release from hydrogels, with physical methods generally correlating to more rapid release relative to chemical conjugation.

A recently explored alternative to this is the creation of a hybrid system in which the bioactive compound is sequestered within gel-loaded emulsions.⁷⁰ Carmo and co-workers reported transdermal delivery of the anticoagulant rivaroxaban (RVX) from surfactant-stabilized emulsions that were encapsulated within a polymer-based topical gel. Using a multiparametric optimization approach, the team identified conditions that would afford emulsions ∼100 nm in size that could be stability loaded into propylene glycol hydrogels. To enhance the concentration of RVX that could be loaded into the gels, the thermoresponsive polymer PF127 was added to the emulsions and led to a nearly 2-fold increase in RVX loading amount by increasing gel viscosity. Addition of a transcutaneous permeation enhancer increased skin RVX diffusion rates from 13.97 μg/cm² for emulsions controls to 18.32 μg/cm² for gels. Prothrombin tests confirmed that the delivered RVX maintained its anticoagulation bioactivity following release from the emulsion–gel hybrid material.

Localized Delivery

Designing an effective antithrombotic delivery system requires incorporating mechanisms for clot-targeting specificity. Off-target distribution of anticoagulants or thrombolytics both limits their therapeutic efficacy and increases the risk of adverse events. In this section, we summarize recent approaches to preferentially target nanoscale therapeutics to sites of thrombosis. We focus on nanoparticles since micron-scale carriers have been limited in their development and scope (for reasons discussed earlier), and macroscale vehicles rely on local administration or transdermal release, thereby limiting the importance of clot-specific targeting.

Physical Stimuli

A materials-centric approach seeks to exploit the unique physical properties of the nanocarrier to enable preferential clot localization. A recent example of this is the physical ferromagnetic behavior of iron oxide nanoparticles, which can be guided to accumulate at clotting sites via an exogenously applied magnetic field. Using a peristaltic flow loop (Figure 6), Prilepskii et al. showed that magnetic guidance of iron oxide particles loaded with the thrombolytic uPA improved the clot removal efficiency of the system by nearly 2-fold compared to unguided samples.⁵¹ Specifically, the time to reduce the clot size by 60% was 7 min for magnetically guided particles, 12 min for nanoparticles without magnetic targeting, and 35 min for free uPA.

Figure 6.

(A) Schematic of magnetic clot-targeting experimental setup, consisting of a peristaltic pump, magnetic stirrer with a water bath, and optical microscope. An enlarged microscopic view is shown inside the red circle. (B) Photo of clot embedded within silicon tubing (dashed lines). Magnetic nanoparticles (MNPs) can be visualized as dark-brown areas within the pale-brown fibrin clot. Reproduced with permission from ref (51). Copyright 2018 American Chemical Society.

In another example, Chen et al. utilized the high shear stress region associated with thrombi formation to design mechanically actuated heparin delivery vectors.⁵⁸ Here, heparin was conjugated to PLL through disulfide linkages and coassembled to form a nanoparticle. The nanocarrier was then electrostatically adsorbed to the surface of RBCs. At an elevated shear stress of 10 Pa, analogous to the site of the thrombus, over 50% of the heparin–PLL nanoparticles detached from RBCs over a 24 h period. Conversely, at a 1 Pa shear pressure, mimicking that of normal flow, 20% of the nanoparticles had detached from the RBC carrier. Together, these examples highlight the potential of physical stimuli–magnetism for iron oxide particles, and of fluid shear force for RBCs, to enable triggered delivery of the therapeutic to sites of thrombosis.

Active Targeting

An analogous strategy to triggerable delivery mechanisms is active targeting via specific molecular interactions. Active strategies for localized clot delivery use biochemical targeting of various motifs specific to, or overexpressed within, growing thrombi, including activated clotting factors, fibrin proteins, and platelets. Peptides represent a privileged class of such agents, as they possess high affinity for their target, can be chemically modified to improve stability, and are readily synthesized through accessible chemical methods.⁷¹⁻⁷³ Among the multitude of peptide ligands available, RGD has been most frequently employed for its binding specificity to α_IIbβ₃ integrins overexpressed on platelets during thrombotic activation.^46,57 Liposomes functionalized with cyclic RGD (cRGD), for example, showed a 10-fold improvement in platelet binding relative to nontargeted liposomal controls.⁴⁶ This ultimately yielded a tripling of the clot dissolving efficacy for tPA-loaded cRGD liposomes compared to controls. CREKA is another common pentapeptide that has been exploited for nanoparticle targeting due to its ability to preferentially bind fibrin.^74,75 A recent platform utilizing the CREKA peptide showed that its functionalization to polymeric nanoparticles increased their specific clot-targeting by over 5 times compared to nontargeted formulations.³⁷

Alternative strategies targeting molecules adjacent to the clotting site, rather than the thrombus itself, have also been explored. For instance, Lee et al. modified PLGA nanoparticles with hyaluronic acid (HA) to target CD44 receptors overexpressed on human mesenchymal stem cells during vessel injury.³⁶In vitro models using human bone mesenchymal stem cells (hBMSCs) showed cell binding increased by nearly two times for nanocarriers containing HA relative to those that did not.

Passive Targeting

The converse approach to active targeting is passive targeting. This relies on the shape and size of the nanocarrier itself to direct improved localization to clotting sites. With regard to shape, computational modeling has shown that cylindrical nanoparticles demonstrate improved adhesion kinetics to vascular tissue surfaces versus spherical analogues.^76,77 For instance, under vascular flow conditions the binding probability of a nanorod under a shear rate of 8 s^–1 is three times higher at the vascular wall than that of a similar volume nanosphere.⁷⁶ This difference in activity was attributed to the larger surface area and tumbling motion of rods versus spheres. Similar findings were reported for disk-shaped particles compared to spheres, which was again attributed to their large surface area and tumbling dynamics.⁷⁷ Experimental validation of this came during demonstration of improved clot-targeting for rod-shaped tobacco mosaic virus (TMV) nanoparticles relative to spherical-shaped cowpea mosaic virus (CPMV) carriers at the site of thrombi formation in mice.⁵⁴ Follow up studies reduced this to practice and demonstrated that TMV carriers loaded with the thrombolytic tPA showed a 3-fold enhancement in clot-targeting specificity relative to free tPA.⁵⁵

In addition to shape, particle size can play a large role on the performance of antithrombotic nanotechnologies. However, the selection of an optimal carrier size is more nuanced and requires balancing multiple competing factors. For example, <100 nm nanoparticles are able to efficiently diffuse within, and percolate through, the fibrin clot network, which possesses pores of 1 nm to 1 μm in size.^31,32 Particles >100 nm are also preferentially phagocytosed and cleared by the mononuclear phagocytic system, which predominately encompasses immune cells of the liver, spleen, and bone marrow.⁷⁸ On the other hand, particles smaller than 10 nm are rapidly removed from circulation via kidney filtration, ultimately leading to their excretion in urine.⁷⁹ Thus, 10–100 nm particle sizes are generally preferred, as they both limit renal clearance, improve serum circulation time, and aid in clot-targeting efficiency.

Translational Application

Clinical use of antithrombotics ranges from emergency thrombolysis to prophylactic anticoagulation. Biomaterial design has, therefore, prioritized clot-targeting precision and kinetics to meet the unique delivery needs of each clinical scenario. Here, we summarize the in vivo performance of antithrombotic technologies in the context of their translational utility. While comparisons are drawn, it is worth noting these are qualitative, not quantitative, assessments, as animal models, antithrombotic cargo, and experimental conditions vary widely between each material platform.

Emergency Thrombolysis

Rapid delivery of thrombolytics to reduce clot lysis time is an urgent priority for emergency treatment of strokes, heart attacks, and pulmonary embolisms. As a result, thrombolytic biomaterials are designed to be intravenously administered and preferentially localize to the clotting site to reopen occluded vessels. For instance, Xu et al. showed that functionalizing acrylamide polymeric nanoparticles with the clot-targeting CREKA peptide improved their localization to lung emboli by more than five times, relative to untargeted control particles, in a mouse pulmonary embolism model.³⁷ As a result, delivery of the anticoagulant hirudin was markedly more rapid and efficient, yielding >80% clot removal 15 min after embolism initiation when using the targeted particles and only 25% thrombolysis for untargeted formulations. A similar increase in mouse survival from 50% to 100% was observed because of clot-targeting in this intrapulmonary thrombosis model. The same study went on to explore the impact of particle targeting in an intravenous FeCl₃-induced carotid arterial thrombosis model. As expected, hirudin-loaded particles possessing the CREKA-targeting peptide reduced the embolization rate from ∼75% to <15%, relative to untargeted particles, and better maintained arterial flow velocity times during carotid thrombosis.

Our group has also demonstrated the translational significance of clot-targeting on nanoparticle-mediated thrombolysis.⁵⁷ Here, peptide nanoemulsions possessing RGD-targeting ligands were found to selectively bind activated platelets in the growing thrombi. This enabled preferential accumulation of the nanoemulsions at the surfaces of thromboplastin-induced clots in an ex vivo bovine embolism model. Subsequent acoustic oscillation of the emulsions during ultrasound imaging led to mechanical dissolution of the clot. Thus, this work not only highlights the importance of clot-targeting on therapeutic efficacy but also demonstrates a complementary approach to removing emboli through mechanical, rather than pharmacologic, means.

Interestingly, studies by Prilepskii et al. showed that preferential capture of even bare nanoparticles, without any added targeting agent, is sufficient to improve the efficacy of loaded thrombolytics.⁵¹ Utilizing two in vivo carotid artery embolism models, this group showed that magnetite nanoparticles loaded with urokinase (uPA) reduced reperfusion times from 11 min in rats, and 9.5 min in rabbits, for the free agent, to <1 min for nanoparticle formulations (Figure 7A). Blood flow rates after reperfusion were similarly ∼4 times faster for animals treated with magnetite delivered uPA relative to the free enzyme (Figure 7B). These studies were all done in the absence of magnetic localization to clots, demonstrating that passive trapping of circulating nanoparticles by the fibrin clot mesh can enhance kinetic localization of thrombolytics to preformed emboli.

Figure 7.

Restoration of blood flow after i.v. injection of free urokinase (uPA) or uPA loaded onto magnetite nanoparticles (MNPs@uPA). (A) Time to vessel reperfusion. (B) Rate of blood flow 24 h postinjection. No reperfusion was observed in the saline cohort. #p < 0.001; *p < 0.001 compared to uPA cohort. Reproduced with permission from ref (51). Copyright 2018 American Chemical Society.

Similarly, utilizing nonspherical shapes can passively improve the thrombolytic efficiency of nanoparticles.^55,56 For example, rod-shaped TMV nanoparticles developed by Pitek et al. improved passive targeting of tPA to arterial thrombi formed in mice by more than 4 times relative to the free enzyme and resulted in 100% of animals showing arterial reperfusion 2 h after initial carotid embolization.⁵⁵ Moreover, passive targeting was shown to reduce the risk of hemorrhage during thrombolytic treatment, cutting the mean tail bleed time from 800 s for free tPA to 400 s for TMV-tPA formulations. This assay, which utilizes cessation of tail bleeding as a surrogate for hemorrhage risk, nicely highlights the additional advantage of using nanoparticle delivery vehicles to reduce the risk of adverse events during thrombolytic therapy.

Preventative Anticoagulant Therapy

The antithesis of emergency thrombolysis is sustained prophylactic anticoagulation in patients with genetic or acquired hypercoagulable states (e.g., cancer, obesity, HIV, etc.). For these indications the goal is to develop sustained release strategies that prolong active anticoagulant concentrations within the systemic circulation, while avoiding overdosing complications. Nanoparticles, again, have been center stage. For example, carbohydrate nanoparticles loaded with the anticoagulant Enoxaparin (Enox) extended the drug’s circulatory half-life from 5 h to over 144 h.³⁹ Similarly, nanoparticles prepared via disulfide cross-linking between heparin and thiol-modified polylysine extended the therapeutic effects of the anticoagulant from 8 to 36 h in mice.⁵⁸ Additional ligation of the nanoparticles to RBCs further prolonged the heparin release profile to a duration of 72 h. These results exemplify the ability of nanoparticle technologies to extend the therapeutic persistence of anticoagulants from several hours to multiple days.

Despite their advantages, nanoparticle therapeutics still require intravenous administration to be operational. Cutaneous anticoagulant patches are painless alternatives that exploit transdermal diffusion to achieve sustained delivery of antithrombotics to the circulatory system. As an illustrative example, polyvinyl microneedles developed by Men et al. extended the time to the maximum serum concentration of hirudin from 0.5 h when administered subcutaneously to mice to 1.5 h for the transcutaneous microneedle delivery system.⁶⁵ Development of bioresponsive microneedles by Gu and co-workers, which released heparin in response to procoagulant thrombin activation, similarly showed sustained anticoagulant effects in murine models.⁶⁶ Excitingly, this system reduced the mortality rate in a lethal thromboembolism model from 100% for mice treated with intravenous heparin to 0% for animals receiving the thrombin-responsive microneedle patch.

Finally, hydrogel coatings have been developed to prevent coagulation at the surfaces of implanted blood contacting devices. Ex vivo evaluation of polytetrafluoroethylene vascular grafts coated with thrombin-responsive heparin–PEG hydrogels, for example, completely prevented clot formation on the graft surface during a 1 h blood contact time, while control surfaces elicited rapid coagulation under similar conditions.⁶⁸ Similarly, electrostatically cross-linked heparin–collagen hydrogels reduced platelet adhesion by 90% relative to nonheparinized gels.

Translational Challenges

While biomaterials have played a central role in advancing new “smart” antithrombotics, there remain several barriers to their widespread clinical adoption. The first is the cost and complexity of the material building blocks. For example, while most linear polymers are readily scalable, the star-shaped and ligand-functionalized analogues utilized to create bioresponsive material networks can be costly to produce at clinically relevant scales. Incorporation of biological components, such as peptides and/or recombinant proteins, can further drive untenable technology costs and production times. Finally, long-term toxicity concerns must be addressed for materials utilizing synthetic polymers, iron oxide, or viral protein precursors. While tolerance may not be a driving issue during acute application (e.g., thrombolysis), repeat administration during long-term anticoagulation treatments may generate chronic side-effects.

Yet, these translational challenges are not unique to antithrombotic materials, and there are several well-established approaches in biomaterials design suited to address these barriers. For example, new monomers and novel polymerization chemistries being developed are poised to create scalable routes to produce complex polymer precursors. In parallel, biofermentation approaches can enable low-cost production of peptide and protein constituents. Recent development of promiscuous synthetases has now expanded biofermentation capabilities to enable the site-specific incorporation of noncanonical amino acids. Finally, inclusion of degradable linkers and biocompatible surface chemistries may address toxicity concerns that arise during clinical application of the discussed systems. Ultimately, with continued development, it is likely these limitations will be comprehensively addressed, and we expect new biomaterial-based antithrombotics will enter the market sometime in the next ten years.

Conclusions and Outlook

Cardiovascular materials represent a major component of the biomaterials market, valued at over $20 billion. Within this field, several emerging technologies are poised to revolutionize the development of new anticoagulants and thrombolytics, as well as advance antithrombotic coatings for blood-contacting medical devices. As presented in this review, the unique pharmacologic and pharmacokinetic properties of nano-, micro-, and macroscale vehicles can be leveraged to create next-generation therapeutics that will drive innovation in this market. In particular, the small size and chemical flexibility of nanoparticle carriers allows them to diffuse within the porous clot scaffold to target, bind, and modify intrinsic biomacromolecules. Conversely, macroscale platforms, such as transdermal patches and hydrogels, can distally respond to clot formation via engineered interactions with circulating factors to controllably release anticoagulants and thrombolytics as needed. Rationally designing these biomaterials for a specified targeting strategy, cargo release rate, and mechanism, as well as mode of administration, allows these platforms to effectively exert either fast acting thrombolysis or sustained anticoagulation. Continued development of these biomaterial-enabled applications is therefore expected to yield several therapeutic innovations in the treatment of cardiovascular dysfunctions. This includes emergent thrombolysis, relevant for stroke, ischemia, and pulmonary embolisms, as well as approaches to manage persistent hypercoagulation (e.g., deep-vein thrombosis, postsurgery, etc.). Controlled release of antithrombotics from biomaterial coatings will also drive the design of new blood contacting devices. Yet, despite several recent advances, producing scalable solutions that can be easily adopted by clinicians remains a key barrier to the widespread translation of antithrombotic biomaterials. Therefore, an equal distribution of resources and effort is needed between the development of new “designer” materials and the production of simple, well-established, and scalable technologies. This will both yield a thriving research ecosystem to drive innovation and ensure that new treatment approaches are being translated into the clinic to produce meaningful benefits for patients in the short term.

The authors declare no competing financial interest.