DRUG DELIVERY BY CARBON NANOTUBE ARRAYS

Abstract

The invention generally relates to carbon nanotube based drug delivery methods, devices, and compositions. More particularly, the invention relates to controlled drug delivery using anchored carbon nanotube arrays.

Description

TECHNICAL FIELD OF THE INVENTION

The invention generally relates to carbon nanotube based drug delivery methods, devices, and compositions. More particularly, the invention relates to controlled drug delivery using anchored carbon nanotube arrays.

BACKGROUND OF THE INVENTION

Targeted, localized and controlled drug delivery remains a major challenge. In many cases, efficacy of a drug can be improved and the risks of side effects reduced if the therapy is administered locally and/or continuously, rather than through conventional oral ingestion or injection, which produce burst releases. In some cases, dose-limiting toxicity levels are caused by agent losses in vascular travel during transplant procedures. Continuous and accurate local dosing is highly desirable, but remains a major challenge, particularly in the cardiovascular field where requirements for a material's biocompatibility and dosing control are stringent.

Current diffusion-based drug-delivery platforms suffer from very slow mass-transfer process. The published reports indicate involvement of solid/solid diffusion as well as channel (e.g., tubule) and solvent-help (e.g., capillary, osmotic) mechanisms but not convection. (Tepe, et al. 2007 Touch Briefings 2007—Interventional Cardiology, pp. 61-63; Scheller, et al. 2004 Circulation 110:810-814; Diaz, et al. 2005 J Biol Chem 280:3928-3937; Creel, et al. 2000 Circ Res 86:879-884; Lovich, et al. 2001 J Pharm Sci 90:1324-1335; Zilberman, et al. 2008 J Biomed Mater Res 84A:313-323; Davies 1997 N Engl J Med 336:1312-1314; Arakawa, et al. 2002 Arterioscler Thromb Vase Biol 22:1002-1007; Parekh, et al. 1997 Gen Pharmac 29:167-172; Heam, et al. 2009 Nature 458:367-371; Celermajer 2002 European Heart Journal Supplement F:F24-F28; Andersen, et al. 2006 BMC Clinical Pharmacology, published online 13 Jan. 2006; Oreopoulos, et al. 2009 J Structural Biology 168:21-36; Panchagnula, et al. 2004 J Pharm Sci 93:2-177-2183; Migliavacca, et al. 2007 Comput Methods Biomech Biomed Engin 10:63-73; Arifin, et al. 2009 Pharmaceutical Research, published online 29 Jul. 2009.) In percutaneous transluminal angioplasty (PCTA) devices, for example, drug washout and overdose remain serious challenges. For oncology applications, for example, localized delivery of sufficient dose of anti-cancer drug via targeted delivery is highly desirable.

In recent years, carbon nanotubes have attracted attention due to their chemical, mechanical and geometric properties. Carbon nanotubes (CNTs) are allotropes of carbon with a cylindrical nanostructure and are members of the fullerene structural family. Nanotubes are categorized as single-walled nanotubes and multi-walled nanotubes. Carbon nanotubes are strong and stiff materials in terms of tensile strength and elastic modulus respectively. Various techniques have been developed to make nanotubes, such as arc discharge, laser ablation, high-pressure carbon monoxide, and chemical vapor deposition.

Researches have been reported on CNT-based drug delivery. For example, a recent study was reported on drug delivery using PEGylated-CNTs. (Liu, et al. 2008 Cancer Res. 68: (16), 6652). The reported system is based on covalently attaching drug molecules to PEGylated CNTs. Another research group used carbon nanotube-based tumor-targeted drug delivery system, which consisted of a functionalized CNTs linked to tumor-targeting modules as well as prodrug modules. (Chen, et al. 2008 J Am Chem Soc 130:16778-16785.) In both of the afore-mentioned approaches, functionalization of the CNTs is required, which presents a number of complications and procedural drawbacks.

One reported example of angioplasty drug delivery is a PTCA balloon coated with paclitaxel in an iopromide matrix. The balloon is inflated for 30-second contact with vascular wall to allow the matrix to dissolve and paclitaxel to migrate into the smooth muscle cell. (Scheller, et al. 2004 Circulation 110:810-814.) Major problems with this device include iopromide being hydrophilic and an X-ray contrast agent. The first causes some drug loss to blood stream (although claimed to be about 6%) and the second leads to adverse reactions for some patients. Furthermore, the balloon still contains about 10% paclitaxel after detachment and only about 15% remains in the plaque.

Another reported example of angioplasty drug delivery is a system using vascular stents made of paclitaxel-eluting composite fibers to deliver about 40% of drug, most of it over 30 days. (Zilberman, et al. 2008 J Biomed Mater Res 84A:313-323.) Since the main mass-transfer mechanism of this device is diffusion, the rate is inherently slow. These drawbacks are in addition to the well-documented risks and side effects associated with stents.

For angioplasty drug delivery monitoring, existing technologies typically use a fluorescent dye administered intravenously through a central venous line with a dose adapted to body weight. (Detter, et al. 2007 Circulation 116:1007-1014; Hattori, et al. 2009 Circ Cardiovasc Imaging 2:277-278; Hosono, et al. 2010 Interact CardioVasc Thorac Surg 10:476-477; Tanaka, et al. 2009 J Thorac Cardiovasc Surg 138:133-140; Waseda, et al. 2009 JACC Cardiovascular Imaging 2:604-612.). The illumination is provided by near-infrared laser diodes with a typical output of 80 mW in a field of view of 10 cm in diameter, eliminating tissue warming and eye protection concerns. The fluorescence emission of the excited dye is typically detected by an IR-CCD camera and digitized with a frame grabber that provides real-time recording.

Therefore, there remains an urgent and unmet need for improved drug delivery systems addressing the above-mentioned shortcomings, particularly in the field of angioplasty drug delivery.

SUMMARY OF THE INVENTION

The invention is based, in part, on the unique approach to drug delivery using anchored carbon nanotube arrays. In particular, the invention provides targeted, localized, and controlled drug delivery using novel acnhored carbon nanotube arrays that carry (e.g., non-covalently) the agent to be delivered, including therapeutic and diagnostic agents.

In one aspect, the invention generally relates a method for delivering an agent to a patient in situ. The method includes: (a) providing a plurality of carbon nanotubes; (b) depositing the agent to the plurality of carbon nanotubes such that the agent is non-covalenty associated with the plurality of carbon nanotubes; (c) placing the plurality of carbon nanotubes deposited with the agent at a target location in the patient's body; and (d) allowing the agent to diffuse from the plurality of carbon nanotubes, thereby delivering the agent in situ.

In another aspect, the invention generally relates to an implantable drug delivery device. The implantable drug delivery device includes: (a) an implantable device; (b) an array of carbon nanotubes anchored on the implantable drug delivery device; and (c) an agent deposited on the array of carbon nanotubes, wherein the agent is not covalently bound to the carbon nanotubes.

In yet another aspect, the invention generally relates to a method for monitoring in situ delivery of an agent. The method includes: (a) providing a plurality of carbon nanotubes non-covalently associated thereon a pharmaceutical agent and a second agent capable of exhibiting a spatially detectable signal; (b) placing the plurality of carbon nanotubes at a target location in the patient's body; and (c) measuring the detectable signal exhibited from the second agent to monitor the delivery of the pharmaceutical agent in situ.

The invention disclosed herein enables improved devices, methods and compositions for treating a number of conditions where targeted, localized, and controlled drug delivery are required.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGS. 1( a)-(b) show an exemplary angioplasty-balloon fitted with carbon nanotubes. FIG. 1(c) and FIG. 1(d) are enlarged views of the carbon nanotubes.

FIG. 2 shows an exemplary carbon nanotube array prepared via capillography methods.

FIG. 3 shows schematics of an array of anchored carbon nanotubes (as tall as 1 mm) on a flexible substrate. (Sansom, et al. 2008 Nanotechnology 19:035302; Sansom 2007 Ph.D. Thesis, California Institute of Technology; Noca, et al. U.S. Pat. No. 7,491,628; Huang, et al. 2007 Nanotechnology 18:305302; Gharib, et al. U.S. Pat. Appl. 20080145616A1.)

FIG. 4 shows exemplary surface-to-volume ratio comparisons of bundled anchored NTs and conventional microneedles. For the same total needle diameter of 50 μm, a bundle of ACNT has almost 100 times larger surface area than a conventional microneedle, assuming a modest CNT spacing of 100 nm center-to-center. The uppermost line represents the surface-to-volume gain by decreasing the CNT spacing by a factor of 2.

FIG. 5 shows a schematic drawing of anchored CNTs (as tall as 1 mm) with drugs deposited in its interstices. (Zhou, et al. 2006 Nanotechnology 17:4845-4853.)

FIG. 6( a) shows a flutax-coated anchored CNT array. FIG. 6(b) shows an uncoated CNT array.

FIG. 7( a) shows exemplary results of recovered fluorescein sodium and flutax-1 with the same initial concentration in DI water. FIG. 7(b) shows exemplary results of recovered fluorescein sodium with various initial concentrations in DI water.

FIGS. 8( a)-8(b) show a rubbing experiment using 3×1×5/128-inch glass microscopy slide rubbed on substrates made of CNT-fitted latex sheet (FIG. 8(a)) and bare latex sheet (FIG. 8(b)).

FIGS. 9( a)-(b) are exemplary images of 3×1×5/128-inch glass microscopy slides used in rubbing experiment. A significant amount of flutax-1 has been washed-out from the substrate made of bare latex sheet in FIG. 9(a) and almost no flutax-1 washed-out from a substrate made of CNT-fitted latex sheet in FIG. 9(b).

FIG. 10 shows exemplary recovered flutax-1 in pure ethanol after the rubbing experiment, where the bare latex sheet shows that no more flutax-1 remains (left) while much flutax-1 still remains on the CNT-fitted latex sheet (right); tubes OD=16 mm.

FIG. 11( a) shows an exemplary PET angioplasty balloon without CNTs. FIG. 11(b) shows a CNT-partially-fitted PET angioplasty balloon. The black region on the CNT-partially-fitted PET angioplasty balloon is an array of vertically aligned CNTs anchored on latex, which cover the outer surface of the PET balloon.

FIGS. 12( a)-(b) show an exemplary experiment setup using agar gel inside a 860-μm-ID, 1,450-μm-OD glass capillary tube with a substrate made of CNT-fitted latex sheet (FIG. 12(a)) and bare latex sheet (FIG. 12(b)).

FIGS. 13( a)-(f) are an exemplary series of time lapse images of mass transfer in 5 weight % agar gel, where the flutax-1 were transferred via ˜300 μm CNT-fitted latex sheet (left) and bare latex sheet (right).

FIGS. 14( a)-(b) show an exemplary mass transfer profile of flutax-1 in 5 weight % agar gel, where the flutax-1 were transferred via ˜300 μm CNT-fitted latex sheet (FIG. 14(a)) and bare latex sheet (FIG. 14(b)).

FIG. 15( a) shows a graph of location of 1% concentration vs. time for a dried flutax-1 positioned as a shallow flat layer (squares, smooth line) and within and anchored CNT array (triangles, dashed line). The squares and triangles mark experimental data and the dashed and smooth lines are numerical solutions of the appropriate model. Schematic geometric descriptions of the problem for a thin flutax layer and CNT array are presented in FIG. 15(b) and FIG. 15(c), respectively.

DETAILED DESCRIPTION OF THE INVENTION

The invention is based in part on the discovery of a universal drug-carrying and delivery platform capable of delivering targeted, localized, and controlled amounts of drugs to a target location within a patient's body. The devices, systems, methods and compositions of the invention significantly remedy the difficult dose control problems in current PCTA devices as well as the complexity and costs associated with CNT functionalizations. Different from the common methods that have been published earlier (e.g., Tepe, et al. 2007 Touch Briefings 2007—Interventional Cardiology, pp. 61-63; Scheller, et al. 2004 Circulation 110;810-814; Diaz, et al. 2005 J Biol Chem 280:3928-3937; Creel, et al. 2008 Circ Res 86:879-884), the device of the invention incorporates a drug on CNTs without having to functionalize the CNTs. Conjugating drugs with functionalized CNTs is inherently complicated and expensive. Additionally, the functionalizing process renders the CNTs hydrophilic, such that the drug attached to the CNTs is vulnerable to washout by the blood stream. Additionally, as disclosed herein, the ability to penetrate the arterial plaque cap and to deliver the drugs directly to the inner part of the plaque is a major advantage of a CNT-fitted angioplasty balloon over the conventional angioplasty-balloon or stent-fitted angioplasty balloon.

CNTs, since their discovery, have been studied for various applications ranging from highly adhesive layers, heat sinks, structural composites, vibration damping layers, lithium storages, to field emitters, due to their exceptional properties. (Iijima 1991 Nature 354; 56-58; Yurdumakan, et al. 2005 Chem Commun 30:3799-3801; Xu, et al. 2004 IEEE Inter Society Conference on Thermal Phenomena 29:549-555; Veedu, et al. 2006 Nature Materials 5:457-462; Daraio, et al. 2004 Appl Phys Lett 85:5724-5726; Wang, et al. 2006 Metals and Materials Int'l 12:413-416; Manohara, et al. 2005 J Vac Sci Tech B23:157-161). The strength and flexibility of CNTs are just two of many exceptional physical properties of CNTs. The accepted Young's elastic modulus for individual CNTs is extraordinarily high, approaching 1 TPa (somewhat less for multi-walled CNTs than single-walled CNTs), which is almost 5 times higher than stainless steel. (Wong, et al. 1997 Science 277:1971-1975, Krishnan, et al. 1998 Phys Rev B58:14013-14019.)

This high Young's modulus of elasticity allows the anchored CNTs to penetrate soft surfaces such as biological tissues upon touching and consequently have the potential of delivering drugs directly to the inner part of the tissue. According to the linear elastic beam theory, the critical force for bucking of an axially loaded clamped cylindrical beam in compression can be calculated by:

$F_{\mathrm{crit}}=\frac{{\pi }^3{\mathrm{Ed}}^4}{2S6L^2}$

where d is the diameter of the beam, L is the length, and E is the Young's modulus. Assuming a bundle of several CNTs is clamped together on the substrate, and typical diameter and length of a bundle of CNTs are 1 μm and 500 μm, respectively, the critical buckling force for each bundle of CNTs is around 0.484 μN, or equivalent to 617 kPa. This critical buckling force for each bundle of CNTs is great enough to overcome the required fracture stress of the arterial plaque cap, for example, about 254.8 kPa. (Holzapfel, et al. 2004 J Bio Eng 126:657-665.) As it is demonstrated herein, CNT bundles anchored on a flexible substrate are sufficiently strong to penetrate soft tissues, e.g. arterial wall or arterial plaque cap, when pressed upon and can deliver a drug directly to the inner part of these targets.

Another extraordinary property of CNTs that makes them a great fit for drug delivery is the hydrophobicity of CNTs. By placing the drugs in the interstices of an array of CNTs, the highly hydrophobic CNTs protect the drugs from being washed out by any aqueous solution. Therefore, the CNT-fitted angioplasty balloon, for example, would protect the drugs from being washed-out during vascular travel to the target location.

Thus, in one aspect, the invention generally relates a method for delivering an agent to a patient in situ. The method includes: (a) providing a plurality of carbon nanotubes; (b) depositing the agent to the plurality of carbon nanotubes such that the agent is non-covalently associated with the plurality of carbon nanotubes; (c) placing the plurality of carbon nanotubes deposited with the agent at a target location in the patient's body; and (d) allowing the agent to diffuse from the plurality of carbon nanotubes, thereby delivering the agent in situ.

In another aspect, the invention generally related to an implantable drug delivery device. The implantable drug delivery device includes: (a) an implantable device; (b) an array of carbon nanotubes anchored on the implantable device; and (c) an agent deposited on the array of carbon nanotubes, wherein the agent is not covalently bound to the carbon nanotubes.

In yet another aspect, the invention generally relates to a method for monitoring in situ delivery of an agent. The method includes: (a) providing a plurality of carbon nanotubes non-covalently associated thereon a pharmaceutical agent and a second agent capable of exhibiting a spatially detectable signal; (b) placing the plurality of carbon nanotubes at a target location in the patient's body; and (c) measuring the detectable signal exhibited from the second agent to monitor the deliver of the pharmaceutical agent in situ.

In certain embodiments, the agent is a pharmaceutical agent capable of providing a therapeutic effect on the patient. Exemplary pharmaceutical agents include: adrenaline (epinephrine), amphetamine, atropine taxol (or its fluorescent derivative flutax-1), and statins.

In certain other embodiments, the agent is a diagnostic agent capable of providing a detectable signal or image indicating a biologically relevant state of the subject. Exemplary diagnostic agents include: electrochemical detectors of chemical warfare agents, utilizing superior electrical properties of carbon nanotubes.

In certain preferred embodiments, the agent comprises an aromatic moiety (e.g., arenes and heteroarenes). The aromatic moiety may include one or more heteroatoms selected from N, O and S. The aromatic moiety may include a single aromatic ring or two or more independent or fused rings. The agent may be mono-aromatic or multi-aromatic.

In certain embodiments, the agent comprises a non-aromatic extended π-bond system.

Preferably, the agent may have a molecular weight from about 130 to about 1500 (e.g., from about 130 to about 1200, from about 130 to about 1000, from about 200 to about 1500, from about 200 to about 1200, from about 200 to about 1000, from about 500 to about 1000).

In certain preferred embodiments, the plurality of carbon nanotubes are in an array format and anchored on an implantable device. In certain preferred embodiments, the plurality of carbon nanotubes are not surface functionalized, e.g., for covalent attachment.

In certain preferred embodiments, the implantable device is an angioplasty balloon.

The angioplasty balloon is preferably anchored uniformly with the plurality of carbon nanotubes, e.g., without surface functionalization. The density of the plurality of carbon nanotubes is typically from about 10¹⁰ nanotubes/cm² to about 10¹¹ nanotubes/cm² (e.g., about 2×10¹⁰ nanotubes/cm², 4×10¹⁰ nanotubes/cm², 6×10¹⁰ nanotubes/cm², 8×10¹⁰ nanotubes/cm².) (FIGS. 1(a)-(d)).

In a preferred embodiment, the agent is an agent for treating arterial plaque and the target location is interior wall of a vascular lumen of the patient. In some embodiments, the target location is inside an arterial plaque whereby at least some of the carbon nanotubes penetrate the outer lining and enter the interior of an arterial plaque.

The carbon nanotubes may be prepared by my suitable means, e.g., by thermal chemical vapor deposition. In some embodiments, the carbon nanotubes are multi-walled. In some embodiments, the carbon nanotubes are single-walled. The desired specifications of carbon nanotubes are dependent on the applications including the requisite strength of the nanotubes.

In certain embodiments, the agent is deposited to the carbon nanotubes by a method that includes: mixing the agent in a low surface tension solvent forming a solution of the agent; coating the plurality of carbon nanotubes with the solution of the agent; and drying the coated plurality of carbon nanotubes, thereby removing the solvent while leaving the agent associated with the plurality of carbon nanotubes.

The low surface tension solvent may be any suitable solvent, for example, an alcohol. In certain preferred embodiments, the low surface tension solvent is pure (200 proof) ethanol or another water-free alcohol.

The drug load is determined according to the requirements of the application. Drug load may range from about 0.1 mg to about 20 mg (e.g., from about 0.1 mg to about 10 mg, from about 0.1 mg to about 5 mg, from about 0.1 mg to about 2.5 mg, from about 1.0 mg to about 20 mg, from about 1.0 mg to about 10 mg.)

CNTs can be formed in several ways, a simple way being a CVD method with a catalyst-coated substrate (typically a few nanometers of iron coated onto silicon wafer) prepared in advance and placed in a tube furnace under saturated flow of carbon containing feed-gas (e.g., ethylene) small portion of reducer feed-gas, e.g. hydrogen, and elevated to proper temperature (e.g., 725° C.). Thermal CVD growth of CNTs in this way generates vertically aligned CNTs on the growth substrate. (Sansom, et al. 2008 Nanotechnology 19:035302; Sansom 2007 Experimental Investigation on Patterning of Anchored and Unanchored Aligned Carbon Nanotube Mats by Fluid Immersion and Evaporation, Ph.D. Thesis, California Institute of Technology.)

Various methods and techniques have been developed that allow the preparation of anchored CNTs, including microscale fluid transport and control techniques by “nanowicking” and by self-assembly pattern formation and a method for controllable anchoring of CNTs within polymer layers (Sansom, et al. 2008 Nanotechnology 19:035302; Sansom 2007 Experimental Investigation on Patterning of Anchored and Unanchored Aligned Carbon Nanotube Mats by Fluid Immersion and Evaporation, Ph.D. Thesis, California Institute of Technology; Zhou, et al. 2006 Nanotechnology 17:4845-4853; U.S. Pat. No. 7,491,628 Noca, et al.; Huang, et al. 2007 Nanotechnology 18:305301; U.S. Pat. App. 20080145616A1 by Gharib, et al.). Tall CNT arrays have been grown in arbitrary patterns (e.g., bundles, rows, geometric shapes) on a substrate with heights of over 1 mm (Bronikowski 2007 J Phys Chem C 111:17705-17712.)

In contrast to the length of CNTs that can be varied relatively easily by varying either the thickness of catalyst layer or the growth time, the packing density of an array of CNTs is relatively harder to vary. One way to adjust the packing density of the CNTs is by changing the timing and duration of the hydrogen exposure during the CNT growth. (Nessim, et al. 2008 Nano Lett 8:3587-3593.) By this method, the packing density of as-grown CNTs can be varied from 3.9×10⁹ to 4.9×10¹⁰ CNTs/cm². Another approach is by compressing the as-grown CNTs by external mechanical force or capillary force. (Wardle, et al. 2008 Adv Mater 20:2707-2714; Futaba, et al. 2006 Nature Material 5:987-994). By using an external mechanical force to compact the as-grown CNTs, the packing density of CNTs can be increased up to about 20%. By using a capillary force to collapse a pack of CNTs, the packing density of CNTs can be increased up to about 50%, or equivalent to the increase of packing density from 4.3×10¹¹ to 8.3×10¹² CNTs/cm².

A major challenge for the vertically aligned CNTs is the poor adhesion of the CNTs to their growth substrates, e.g., silicon wafer. Although the mechanical properties of individual CNTs are excellent, the collective properties of bulk vertically-aligned CNTs have not been as good as expected, because of the weak bond between the base of the CNTs and their growth substrates. A method has been developed to overcome this problem by anchoring the CNTs in a layer of flexible polymeric materials, e.g., PDMS (polydimethylsiloxane), PMMA (poly(methyl methacrylate)), or latex, (Sansom, et al. 2008 Nanotechnology 19:035302.) As-grown CNTs are manipulated by handling their growth substrate, inverted into a spin-coated polymer layer, and the assembly is cured (usually by heat). (FIG. 2).

The CNT's growth substrate may then be removed. The CNT-growth can be patterned by controlling the patterning of the thin iron catalyst layer prior to the CNT-growth step. Through separate control of polymer layer thickness and CNTs length, the depth of anchoring of the CNTs into the flexible polymer layer can be controlled. Thus, any pattern of as-grown CNTs, defined by catalyst pattern may be inverted and anchored into a polymer layer as required.

CNTs anchored in PDMS can withstand the shear stress up to 230 dyne/cm² and tensile stress up to 64.5 kPa. (Sansom, et al. 2008 Nanotechnology 19:035302.). The effective adhesion strength between the CNTs and the flexible polymeric layer depends on the depth of anchoring of she CNTs inks the polymer layer, the strength of the bond between CNTs and the polymer layer, and the fracture toughness of the polymeric layer.

Anchored CNTs on a flexible polymeric layer can be designed and prepared such that they have the requisite mechanical strength ensuring that no CNTs would enter the blood circulation or be left in the living tissues that may pose harm to the patient. The biocompatibility of CNTs has been demonstrated, including evidences that show various types of living ceils (e.g. neuronal cells, osteoblast cells and fibroblast cells) can be supported by CNTs (Hu, et al. 2004 Nano Lett 4:507-551; McKenzie, et al. 2004 Biomaterials 25:1309-1317; Webster, et al. 2004 Nanotech 15:48-54; Gabay, et al. 2005 Physica A 350:611-621; Price, et al. 2005 Biomaterials 24:1877-1887; Elias, et al. 2002 Biomaterials 23:3279-3287; Correa-Duarte, et al. 2004 Nano Lett 4:2233-2236.) Some studies have been reported that CNTs may present certain health problems, especially related to lung toxicity, cytotoxicity, and skin irritation. (Huczko, et al. 2001 Fullerene Sci Tech 9:247-250; Huczko, et al. 2001 Fullerene Sci Tech 9:251-254; Huczko, et al. 2005 Fullerene Nanotubes Carbon Nanostruct 13:141-145; Lam; et al. 2004 Toxicol Sci 77:126-134; Warheit, et al. 2004 Toxicol Sci 77:117-125: Muller, et al. 2005 Toxicol App Pharmacol 207:221-231; Shvedova; et al. 2003 J Toxicol Environ Health A 66:1909-1926; Monteiro-Rivtere, et al. 2005 Toxicol Lett 155:377-384; Jia, et al. 2005 Environ Sci Technol 39:1378-1383; Cui, et al. 2005 Toxicol Lett 155:73-85; Tamura, et al. 2004 Key Eng Mater 254-6:919-922.) Thus, to minimize the risk of side effects, the minimum mechanical strength of the ACNT on a CNTs-fitted drug delivery system must be sufficiently higher than the maximum shear stress induced by the tissue.

Diffusion is an important mechanism in delivering drugs from any pharmaceutical system. The release kinetics of a drug delivery system depends on the initial concentration of the drug and the surface area of the system. Therefore, to improve the drug delivery mechanism, it is crucial for a system to have a sufficiently large surface area. Such a system can essentially be made by fitting an array of anchored CNT arrays onto a flexible substrate (FIG. 3).

The following expression compares the surface area of a CNT-fitted flexible substrate with a bare flexible substrate without CNT-enhancement.

$\frac{A_{\mathrm{CNT}\text{-}\mathrm{fitted}\text{-}\mathrm{substrate}}}{A_{\mathrm{bare}\text{-}\mathrm{substrate}}}=\frac{\frac{2\sqrt{3}\pi \mathrm{dhWL}}{3s^2}}{\mathrm{WL}}=\frac{2\sqrt{3}\pi \mathrm{dh}}{3s^2}$

Where d is the diameter of individual CNTs, h is the length of individual CNTs, s is the distance between individual CNT, W and L are the width and length of the flexible substrate respectively. Using the following assumed typical values, where d=10 nm, s=100 nm, and h=500 μm, the following result is obtained:

$\frac{A_{\mathrm{CNT}\text{-}\mathrm{fitted}\text{-}\mathrm{substrate}}}{A_{\mathrm{bare}\text{-}\mathrm{substrate}}}=1813.17$

This indicates that for a CNT-fitted drug delivery platform, the mass-transfer surface for delivering drugs is about 1,800 times greater than that of a non-CNT-fitted platform of similar geometry. Therefore, there is a substantial increase in drug mass transfer rate when a drug delivery platform is fitted with CNTs.

Compared to a simple hollow microneedle, the surface-to-volume ratio of an anchored CNT bundle is at least one order of magnitude greater (FIG. 4). This is again crucial for a local drug delivery platform. This large surface-to-volume ratio also shows that a high dose of drugs can be placed in a small size of CNT-fitted drug delivery platform.

The drug depositing technique for depositing drugs to CNTs may be any suitable method. A novel approach disclosed here is to use a very low surface tension liquid (e.g., substantially lower than 72 dyne/cm (water, 25° Celsius), such as isopropanol, ethanol and acetone) to place the drugs on the CNTs. Fluid transport based on wicking through a nano-fibrous material has been previously studied and characterized. (Zhou, et al. 2006 Nanotechnology 17:4845-4853.) The same phenomenon is employed here for depositing the drugs in the interstices of the anchored CNT array from their solutions followed by thorough drying (FIG. 5). Using this technique, anchored CNTs were coated with both flutax-1 and uranine (sodium salt of fluorescein) (FIGS. 6(a)-(b)).

The drug depositing technique disclosed herein is a straightforward and less expensive method to place drugs within an anchored CNT array. Using this method, drugs can be placed in an anchored CNT array without having the need to functionalize the surface of the CNTs. Generally speaking, this method works preferably for drugs that have one or more aromatic moieties and/or extended π-bond systems such as atropine and amphetamine, which help with the creation π-π interactions with CNTs. by creating strong π-π interactions, the need to functionalize the CNT beforehand is eliminated. Reduced surface tension liquids (optionally having a surfactant such as sodium dodecyl sulfate (SDS) and similar detergents) easily wicked through CNT arrays. For example, a device can be deposited with the drugs in the interstices of CNTS using an ethanol solution of the drug followed by thorough drying to remove the solvent.

By attaching a flourescent molecule to the drug, e.g., taxol, direct photographic observation and monitoring of its movement into plaque can be accomplished, for example, using fiber-optic-based technology and calibration methods. (U.S. Pat. No. 5,116,317 by Carson, et al.; U.S. Pat. No. 4,842,390 by Sottini, et al.; Tepe. et al. 2007 Touch Briefings 2007-Interventional Cardiology, pp. 61-63; Scheller, et al. 2004 Circulation 110:810-814.) Substantial increases in drug mass tranfer rate can be achivied when the angioplasty balloon is fitted with anchored CNTs, as disclosed herein.

In one embodiment, flutax-1 (i.e., a conjugated paclitaxel and fluourecein) is used. (Diaz, et al. 2005 J Biol Chem 280:3928-3937; Creel, et al. 2000 Circ Res 86:876-884; Lovich, et al. 2001 J Pharm Sci 90:1324-1335.) Using flutax-1 allows measurement of pclitaxel delivery to an arterial-plaque, optically and directly. This eliminates the need for currently popular high-performance liquid chromotography (HPLC) approach that would need biopsy, or for ³H radio-labeled paclitaxel measurement. (Creel, et al. 2000 Circ Res 86:879-884.) Fabrication of this device is translatable to industrially scalable processes like toll -to-toll manufacturing for combining CNTs on substrates and plymer layers that allows for very low-cost production.

The studies disclosed herein include micromechanics and nano-dynamics relevant to insertion mechanism of nanostructures into model tissues and directly-on-target interstitial mass-transfer phenomena involved with such insertion. The ability to penetrate the arterial plaque cap and deliver the drugs directly inside the plaque is clearly an outstanding advantage of a CNT-fitted angioplasty balloon over the conventional angioplasty-balloon or stent-fitted angioplasty balloon.

Other studies disclosed herein relate to the prevention of the carried drugs from being rubbed-out/washed-away during procedure/travel to the target location. It is important to note that, due to the hydrophobic nature of CNTs, aqueous media such as blood plasma do not easily reach the interior spaces of the CNTs where the drug. Is deposited. By minimizing drug loss, the risks of overdosing patients may be reduced considerably, another outstanding advantage of an anchored CNT-fitted angioplasty over other drug-delivery devices. Studies and results disclosed herein provide critical information needed tor fabricating specific-target drug delivery platforms and for guiding the development of novel drug delivery systems requiring access to interstital spaces.

Drug protection experiments were conducted to show that the CNTs protect the deposited drugs from being washed away by blood-like liquids. Two types of dyes were used in these experiments, the fluorescein sodium that represents the hydrophilic drugs and flutax-1 that represents the hydrophobic drugs. Two types of specimens were used, a bare sheet of latex and a sheet of latex with CNTs anchored on one side. Since the CNTs are highly hydrophobic, both fluorescein sodium and flutax-1 were dissolved in ethanol so that both dyes could go into the interstices of the CNTs. The same amounts of both fluorescein sodium and flutax-1 were put in each specimen and allowed to dry. After the dye dried, each specimen was placed in 2 mL of DI water (to simulate blood) and the concentration of the dye was analyzed using spectrophotometer.

The results showed that the hydrophobicity of CNTs protected the deposited drugs, which could not be easily washed out by DI water. On the contrary, the DI water was able to wash out the fluorescein sodium easily (FIG. 7(a)). Notice the obvious difference of the recovered amount of fluorescein sodium between the bare latex samples and the CNT-fitted latex samples. The bare latex samples lost almost all of their initial concentration of fluorescein sodium, while the CNT-fitted latex samples retained about 50% of their initial concentration of the fluorescein sodium.

Another set of experiments was done using various concentrations of fluorescein sodium on both types of specimens (FIG. 7 (b)). The time to wash out the fluorescein sodium in DI water for the CNT-fitted latex samples depended on the initial concentration. Lower initial concentration lead to longer time that was needed to wash out the fluorescein sodium. This was not observed with the bare latex samples. The bare latex samples could not retain the fluorescein sodium from being washed away by DI water for whatever the initial concentration was.

Another experiment was performed that targeted possible drug removal from CNT-fitted drug delivery platform by mechanical (e.g., rubbing) actions during its cardiovascular travel. Two types of specimens were used, a bare sheet of latex and a sheet of latex with CNTs anchored on one side. On each specimen, approximately 125 μg of flutax-1 was deposited by successively dropping small volumes of its ethanol solution and allowed to dry. Next, 1 μL of DI water followed by a microscopic 3×1×5/128-inch glass slide was put on each specimen. Finally, with a 500 g weight on top of it, the microscopic slide was moved back and forth a few millimeters at about 1 cm/sec for 60 seconds (FIGS. 8(a)-(b)). The results showed considerably higher flutax-1 removal from the bare latex compared to CNT-anchored one (FIGS. 9(a)-(b)). After the test, each of the two specimens was dropped in 2 mL of pure ethanol to see if they still had flutax-1 left on them. There was considerably more flutax-1 released (present) from the CNT-anchored latex compared to the bare one (FIG. 10).

A CNT-fitted angioplasty balloon was prepared with vertically aligned CNT arrays and anchored on a polymeric layer. We started with a 20 mm-long PET angioplasty balloon (Advance Polymers, Item No. 03002016AA). This balloon was then dipped into a liquid latex compound such that the entire surface of the balloon was covered uniformly with a thin layer of latex. Before the latex layer was cured, a silicon substrate, which has as-grown CNTs on it, was placed in the inverted position onto the latex layer. After the latex layer was cured, the silicon substrate was then removed, leaving the CNTs anchored firmly on the latex layer (FIGS. 11(a)-(b)). The right length and packing density of CNTs that allow optimized drugs delivery and protection can be selected and used according to specific requirements of a particular application.

For continuous monitoring of angioplasty drug delivery, fluorescent-conjugated versions of the drugs intended for delivery, for example, flutax-1 instead of taxol, can be used. This method eliminates the need for potentially hazardous administration of a fluorescent dye intravenously through a central venous line. Illuminating can be done with a light source to accommodate the excitation of delivered fluorescent-conjugated drug. A digital camera can be used to record the imagery of emitted light. (Detter, et al. 2007 Circulation 116:1007-1014; Hattori, et al. 2009 Circ Cardiovasc Imaging 2:277-278; Hosono, et al. 2010 Interact CardioVasc Thorac Surg 10:476-477; Tanaka, et al. 2009 J Thorac Cardiovasc Surg 138:133-140; Waseda, et al. 2009 JACC Cardiovascular Imaging 2:604-612.)

The following studies apply Fick's second law of diffusion in a semi-infinite medium, x≧0. The concentration at the applicator (angioplasty balloon, etc.) interface is assumed constant in the first version, diminishing in the second. The first refers to the initial time-span when applicator is still touching the plaque, the second refers to post applicator removal.

One-Dimensional Semi-Infinite Slab Model of Solid-Solid Diffusion, Version I

This model assumes constant concentration at the applicator (angio balloon, etc) interface, (Perry, et al. 1963 Chemical Engineers' Handbook, 4^th edition, McGraw-Hill, Now York.)

$\begin{array}{cc}\frac{\partial C\left(x,t\right)}{\partial t}=D\frac{{\partial }^2C\left(x,t\right)}{\partial x^2}\mathrm{Fick}\text{'}s\mathrm{second}\mathrm{law}\mathrm{of}\mathrm{diffusion}\mathrm{in}a\mathrm{semi}\text{-}\mathrm{infinite}\mathrm{medium},x\ge 0.& \left(1\right)\end{array}$

Simplified Initial and Boundary Conditions

C(x,0)=0   initial condition (2)

c(0,t)=C₀   boundary condition (3)

Simplified Analytical Solution

Laplace transform with respect to t yields,

$\begin{array}{cc}{\int }_0^{\infty }e^{-\mathrm{st}}\frac{\partial C\left(x,t\right)}{\partial t}t=D{\int }_0^{\infty }e^{-\mathrm{st}}\frac{{\partial }^2C\left(x,t\right)}{\partial x^2}t\Rightarrow \frac{{}^2F}{x^2}-\frac{s}{D}F=0\mathrm{because}\mathrm{of}\mathrm{the}\mathrm{initial}\mathrm{condition}& \left(4\right)\\ F\left(x,s\right)=\left(c_0/s\right)e^{-\mathrm{sx}/D},F\left(0,s\right)=c_0/s\mathrm{because}\mathrm{of}\mathrm{the}\mathrm{boundary}\mathrm{condition}& \left(5\right)\end{array}$

Reverse transform yields,

$\begin{array}{cc}C\left(x,t\right)=C_0\left[1-\frac{2}{\sqrt{\pi }}{\int }_0^{x/\left(2\sqrt{\mathrm{Dt}}\right)}e^{-u^2}u\right]\mathrm{for}\mathrm{the}\mathrm{concentration}\mathrm{peak}\mathrm{onwards}& \left(6\right)\end{array}$

One-Dimensional Semi-Infinite Slab Model of Solid-solid Diffusion, Version II

This model assumes diminishing concentration at the applicator (angio balloon, etc) interface. (Mehrer 2007 Diffusion in Solids: Fundamentals, Methods, Materials, Diffusion-Controlled Processes, Springer-Verlag, Berlin.)

$\begin{array}{cc}\frac{\partial C\left(x,t\right)}{\partial t}=D\frac{{\partial }^2C\left(x,t\right)}{\partial x^2}\mathrm{Fick}\text{'}s\mathrm{second}\mathrm{law}\mathrm{of}\mathrm{diffusion}\mathrm{in}a\mathrm{semi}\text{-}\mathrm{infinite}\mathrm{medium},x\ge 0.& \left(7\right)\end{array}$

Simplified Initial and Boundary Conditions

C(x,0)=Mδ(x)   initial condition (8)

where M is the number of diffusing particles per unit area and O(x) the Dirac delta function.

$\begin{array}{cc}\frac{\partial C\left(0,t\right)}{\partial x}=0\mathrm{boundary}\mathrm{condition}& \left(9\right)\end{array}$

The analytical solution is the following Gaussian equation

$\begin{array}{cc}C\left(x,t\right)=\frac{M}{\sqrt{\pi \mathrm{Dt}}}\exp \left(-\frac{x^2}{4\mathrm{Dt}}\right)\mathrm{which}\mathrm{at}t=0\mathrm{reduces}\mathrm{to}C\left(x,0\right)=M\delta \left(x\right)& \left(10\right)\end{array}$

The diffusion of drugs from coated anchored CNT arrays and the pressure driven mass transfer through such structures was experimented with diffusion of flutax-1 from an anchored CNT array in agar gels and water-containing protein/lipid/cholesterol structures (including swine and chicken skin). The results indicated that flutax-1 molecules did not move appreciably in DI water, but do so in agar (in a matter of minutes) or gelatin-water-containing matrices. The results with agar gels are presented below (FIGS. 12(a)-14(b)), and fit reasonably with the simple theoretical model. The diffusion from the anchored CNT array was compared with simple diffusion from a thin strip of dried flutax on a bare polymeric sheet (the geometries of both cases examined are illustrated in FIGS. 12(a)-(b).

The diffusion process was modeled with Fick's second law, assuming one-dimensional diffusion and neglecting the melting time of the dried flutax in comparison with the diffusion time within the agar (which is supported by experimental data). Schematic description of the geometry for both examined cases is presented in FIGS. 15(b) and 15(c).

The model for the case of thin layer of flutax (L₁/L1) consists of one-dimensional diffusion within the agar gel

$\frac{\partial C\left(x,t\right)}{\partial t}=D\frac{{\partial }^2C\left(x,t\right)}{\partial x^2},$

and constant concentration C(x=0, t) C₀ and no flux δC (x=0, t)/ox=0 boundary conditions at x=0 and x→∞, respectively. A classical solution for this problem exists in the literature, and is defined as

$C\left(x,t\right)=C_0\left[1-\mathrm{erf}\left(\frac{x}{2\sqrt{\mathrm{Dt}}}\right)\right].$

For the case of flutax positioned within a CNT array (FIG. 15c) L₂ cannot be neglected in comparison with L, and two regions of diffusion are modeled, the agar region and the CNT region. Since the entire length of the CNT array is coated with flutax at t=0, the initial condition is now

$C\left(x,t=0\right)=\{\begin{array}{cc}0,& x>L_2\\ C_0,& x<L_2\end{array},$

Utilizing the one-dimensional case, the reduction in available area for the diffusion process within the CNT region can be modeled by reducing the value of the diffusion coefficient proportionally to the reduction in surface area and the initial concentration C₀ can be estimated from the amount of flutax and the available void space within the CNT array.

FIG. 15( a) presents the numerical solution of the models and the experimental data for thin flutax layer (squares and smooth line) and flutax within CNT array (triangles and dashed line). The general trends of the diffusion are predicted reasonably. The main effect of the CNT array is to change the initial conditions of the diffusion and thus the differences between both cases are expected to decrease as t increase, which is indeed observed in experimental data in accordance with the model.

EXAMPLES

Experimental Methods

CNT Arrays Growth.

The vertically aligned multi-walled carbon nanotube arrays used in this study were grown using thermal chemical vapor deposition on silicon wafer substrates. These wafers were coated with 10 nm aluminum oxide buffer layer and 1 nm iron catalyst layer using electron beam evaporator (Temescal BJD 1800) and diced into 1×1 cm samples. The growth itself was performed in a 1-inch diameter quartz tube furnace (Lindberg/BlueM Single Zone Tube Furnace) under the 490 standard cubic centimeters per minute (seem) ethylene gas (Matheson 99.999%) and 210 seem hydrogen gas (Airgas 99.999%) at a temperature of 750° C. and a pressure of 600 torr. The flow rate and pressure of the gases was maintained by an electronic mass flow controller (MKS πMFC) and a pressure controller (MKS πPC). The overall growth quality, including the length of the array, was characterized under scanning electron microscope (ZEISS LEO 1550VP).

Anchored CNT on Flat Surface.

A thin layer of uncured polymer (e.g., PDMS) was spin-coated (SCS G3 spin coater) onto a flat rigid substrate. The thickness of the polymer layer can be controlled by varying the speed (rpm) and dwell time of the spinner. As-grown carbon nanotube arrays were manipulated by handling their growth substrate, inverted and then inserted into the spin-coated polymer layer. The whole assembly was subsequently cured, usually by heat at elevated temperature (e.g., about 80° Celsius). The CNT's growth substrate could then be easily removed after the polymer layer is fully cured. By controlling the polymer layer thickness and CNTs length, the depth of anchoring of the CNTs into the flexible polymer layer may be controlled.

Anchored CNT on Balloon.

A 20 mm-long PET angioplasty balloon (Advance Polymers 03002016AA) was used as a platform. This balloon was then dipped into a liquid latex compound so that the entire surface of the balloon was covered uniformly by a thin layer of latex. Before the latex layer was cured. CNTs attached to a silicon substrate were inverted then placed into the latex layer. The whole assembly was subsequently cured by leaving it in air at room temperature for 24 hours. After the latex layer was cured, the silicon substrate was then removed, leaving the CNTs anchored firmly on the latex layer on the balloon.

Flutax-1 and Uranine Attachment on CNT.

Two types of dye were used in these experiments: the uranine (sodium fluorescein, Sigma Aldrich 67884) that represents the hydrophilic drugs and flutax-1 (Tocris Bioscience 2226) that represents the hydrophobic drugs. Since the carbon nanotubes are highly hydrophobic, both fluorescein sodium and flutax-1 were dissolved in pure ethanol (Sigma Aldrich E7023), so that both dyes could wick into the interstices of the CNT specimen. The same amounts of both fluorescein sodium and flutax-1 were placed by successively dropping small volumes in each specimen and letting them dry in air at room temperature.

Diffusion Measurement.

Diffusion of flutax-1 from an anchored CNT array was measured in 5% agar gels (Sigma Aldrich 17209) in water. The agar gels were placed inside a 1450 μm OD×860 μm ID glass capillary tube (Clay Adams 4614). The anchored CNT array was then placed flush to the tip of the capillary tube. To determine the diffusion profile of the flutax-1 from the CNT array into the agar gels, time lapsed photographs were taken by fluorescent microscope (Nikon Eclipse TE2000-S). The diffusion profile was then determined from the fluorescence intensity captured in these photographs.

INCORPORATION BY REFERENCE

References and citations to other documents, such as patents, patent applications, patent publications, journals, books, papers, web contents, have been made in this disclosure. All such documents are hereby incorporated herein by reference in their entirety for all purposes.

EQUIVALENTS

The representative examples are intended to help illustrate the invention, and are not intended to, nor should they be construed to, limit the scope of the invention. Indeed, various modifications of the invention and many further embodiments thereof, in addition to those shown and described herein, will become apparent to those skilled in the art from the full contents of this document, including the examples and the references to the scientific and patent literature included herein. The examples contain important additional information, exemplification and guidance that can be adapted to the practice of this invention in its various embodiments and equivalents thereof.

Claims

1. The method of claim 1, wherein depositing the agent to the plurality of carbon nanotubes comprises: (i) mixing the agent in a low surface tension solvent forming a solution of the agent;(ii) coating the plurality of carbon nanotubes with the solution of the agent; and(iii) drying the coated plurality of carbon nanotubes, thereby removing the solvent while leaving the agent associated with the plurality of carbon nano tubes.

2. The method of claim 1, wherein the low surface tension solvent is an alcohol.

3. The method of claim 2, wherein the alcohol is ethanol.

4. An implantable drug delivery device, comprising: (a) an implantable device;(b) an array of carbon nanotubes anchored on the implantable device; and(c) an agent deposited on the array of carbon nanotubes, wherein the agent is not covalently bound to the carbon nanotubes.

5. The implantable drug delivery device of claim 4, wherein the agent is a pharmaceutical agent capable of providing a therapeutic effect.

6. The implantable drug delivery device of claim 4, wherein the agent is a diagnostic agent capable of providing a detectable signal or image indicating a biologically relevant state of the subject.

7. The implantable drug delivery device of claim 4, wherein the agent comprises an aromatic moiety.

8. A method for monitoring in situ delivery of an agent, comprising: (a) providing a plurality of carbon nanotubes non-covalently associated thereon a pharmaceutical agent and a second agent capable of exhibiting a spatially detectable signal;(b) placing the plurality of carbon nanotubes at a target location in the patient's body; and(c) measuring the detectable signal exhibited from the second agent to monitor the deliver of the pharmaceutical agent in situ.

9. The method of claim 8, wherein the agent is a pharmaceutical agent comprises an aromatic moiety.