Small Solutions for Big Problems: The Application of Nanoparticles to Brain Tumor Diagnosis and Therapy

Synopsis

Nanoparticles are likely to play a key role in the future diagnosis and treatment of CNS malignancies. Nanoparticles have the potential to revolutionize both preoperative and intraoperative brain tumor detection. In addition, nanoparticles may also serve as novel, targeted delivery devices for chemotherapy, gene therapy, photodynamic therapy and thermotherapy. The translation of current research in nanotechnology into a clinically relevant component of brain tumor management will rely on solving challenges related to the pharmacology of nanoparticles.

The incidence of primary brain tumors in the United States has been estimated at approximately 43,800 per year (1,2,3). While incidence rates of cancer in general have fallen or been stable, the age-adjusted incidence of primary brain tumors has been increasing at an alarming rate over the past several decades (4,5). Moreover, improvements in surgical and adjuvant therapy for brain tumors have not translated into a meaningful improvement in patient outcome (6). There is a need, therefore to apply novel technology to the treatment of brain tumors. Quantum advances in nanotechnology have the potential to revolutionize multiple aspects of the diagnosis and treatment of brain tumors in the future (7).

Some consider innovations in nanotechnology to be the greatest breakthrough in engineering since the industrial revolution (8). The promise of nanotechnology is based on a rapidly expanding body of knowledge on the synthesis and characterization of ultra-small particles. Nanoparticles (NP) are structures typically with a diameter of 10–200nm (9) that hold virtually limitless possibilities for design and application in biologic systems. Since the application of nanotechnology to the imaging of gliomas was proposed (10), there has been a rapid expansion of the application of nanodevices to the diagnosis and treatment of brain tumors. Nanoparticles are macromolecular structures that can be engineered to carry one or more types compounds to brain tumors including MRI contrast agents, fluorescent and visible dyes, chemotherapeutic agents and photosensitizers. Moreover the efficient, selective delivery of nanoparticles and their payloads to brain tumors can be dramatically enhanced by altering their size and surface chemistry.

This article outlines the rationale and scientific foundation for applying nanotechnology to the diagnosis and management of brain tumors. Current innovations in nanotechnology that hold promise for preoperative and intraoperative imaging are discussed. The potential role for nanoparticles as chemotherapeutic delivery vehicles is also explored. Finally the challenges of translating advances in nanotechnology to the management of brain tumor patients are examined.

Why Nanoparticles?

Numerous physicochemical properties of nanoparticles make them ideal devices for the delivery of specific compounds to brain tumors. Large amounts of small molecules, such as contrast agents or drugs, can be loaded into nanoparticles via a variety of chemical methods including encapsulation, adsorption and covalent linkage. The concept of a single nanoparticle carrying a large number of drug molecules or ions is referred to as “nanoparticle amplification” (11). In addition, multiple types of small molecules with similar or different functions can be incorporated into a given nanoparticle depending on its physicochemical properties (see Figure 1). Moreover, nanoparticle size and chemical composition can be altered to control the efficiency of small molecule loading (12).

Figure 1.

Schematic diagram of multifunctional polyacrylamide nanoparticle. While this nanoparticle contains contrast agents, photosensitizers and F3 targeting peptides, all of these components can be interchanged for other small molecules with various functions. (From Ref 13. Permission requested)

The specificity and efficiency of nanoparticle delivery can be enhanced through the attachment of specific targeting modalities. Multiple targeting molecules can be added to the surface of nanoparticles to improve targeting through a concept referred to as “surface-mediated multivalent affinity effects” (11). In addition, depending on the targeting modality, a certain subpopulation of cells, such as angiogenic cells, may be targeted within a brain tumor (13). Nanoparticle delivery can also be enhanced by adding hydrophilic macromolecules such as polyethylene glycol to the surface of nanoparticles, which alter their half-life and clearance (9).

In contrast to small molecules, which diffuse freely into and out of tissues, nanoparticles behave as macromolecules and therefore have a tendency to accumulate within a tumor via the enhanced permeability and retention (EPR) effect. The EPR effect is thought to result from active angiogenesis, the expression of vascular mediators of extravasation (ie nitric oxide, VEGF and bradykinin), and altered vascular architecture (14). The EPR effect explains, in part, the observation that some nanoparticles are retained within tumor tissue long after their serum levels decline.

Another factor, which allows nanoparticles to serve as efficient vehicles for diagnostic and therapeutic agents, is their ability to isolate their payload from the surrounding environment. For example, methylene blue is rapidly reduced and inactivated by methemoglobin reductase when injected intravenously. However, when loaded into a nanoparticle, methylene blue can be delivered to tumor cells in high concentrations where it can be used as a mediator of photodynamic therapy (15, 16,17). Similarly, the delivery of therapeutic quantities of chemotherapeutic agents to target tumor tissues is often associated with high systemic toxicity. This toxicity can be greatly reduced by encapsulating chemotherapeutic agents into nanoparticles. Therefore, nanoparticles can deliver high concentrations of chemotherapeutic agents specifically to tumor tissue while preventing the undesirable systemic consequences of systemic chemotherapy (18).

Nanoparticle Targeting

A wide variety of nanoparticle targeting options have been reported including peptides, cytokines, drugs, antibodies and ferromagnetic agents. Three peptides have been reported for targeting nanoparticles to surface molecules expressed on glioma cells in vitro and in vivo. Two of the peptides, RGD and F3, were identified for their ability to target cell surface markers on angiogenic epithelium in implanted tumors through the phage display techniques. RGD peptides contain a motif, which binds to αvβ3 integrin, a molecule expressed on the vascular endothelium of implanted murine tumors (19, 20). RGD-targeted nanoparticles have been shown to be effective at delivering near-infrared dye-loaded nanoscale crystals, known as quantum dots to implanted tumors in mice (21). RGD-targeted nanoparticles have also been suggested as chemotheraputic delivery vehicles (22). F3, a 31-amino acid peptide, binds to nucleolin, a cell surface receptor expressed in proliferating angiogenic and tumor cells (20). F3-targeted nanoparticles have been recently shown to be effective at delivering both MRI contrast agents and photosensitizers to implanted brain tumors in rats (13). Chlorotoxin, a peptide isolated from scorpion venom, which targets chloride channels that are upregulated on the surface of glioma cells (23), has also been used to target nanoparticles carrying MRI contrast agents and optical near-IR dyes (24).

In addition, larger peptides such as cytokines and monoclonal antibodies have been used to target nanoparticles to brain tumor cells. For example, IL-13 coated nanovesicles containing doxorubicin have been show to be effective at delivering high drug concentrations to implanted gliomas in mice (25). Monoclonal antibodies, such as anti-epithelial growth factor receptor antibody, have also been recently proposed as a targeting modality for nanoscale drug delivery devices (26). While peptides are the most commonly employed nanoparticle targeting agents, small molecules such as methotrexate have also been used. Methotrexate coated nanoparticles bind to folate receptors, which are upregulated in a variety of human cancers, including gliomas (27). Finally, mechanical targeting of ferromagnetic nanoparticles into brain tumor beds, using external magnetic fields has also been recently suggested (28).

Iron Oxide Nanoparticle-Based MRI Contrast Agents

One of the clearest and most mature applications of nanotechnology to the management of brain tumors is in the area of magnetic resonance imaging. A variety of nanoparticles have been developed as novel MRI contrast agents over the past decade. To date, nanoparticle-based contrast agents have been designed with a core of single or multiple iron-oxide crystals with or without a shell of organic material (9, 29). The organic shell dictates clearance of the nanoparticle by influencing opsonization and endocytosis. For example, clearance of nanoparticle and persistence of contrast within the tumor parenchyma can be delayed in a controlled fashion through the attachment of polyethylene glycol chains of variable length (9, 30).

Both traditional gadolinium-based contrast agents and nanoparticle-based contrast agents cause enhancement of tumors by passing through areas of disrupted blood brain barrier where they alter MR signal intensity. However, once across the blood brain barrier, there are a number of key differences between gadolinium-based contrast agents and iron oxide nanoparticle-based contrast agents. The differences in the way these contrast agents interact with tumor tissue explain why nanoparticle-based iron oxide contrast agents may provide distinct information from traditional MRI contrast agents.

Unlike gadolinium, nanoparticle-based contrast agents, such as ultrasmall superparamagnetic iron oxide (USPIO) have a tendency to be taken up by reactive, phagocytic cells that are commonly found at infiltrating tumor margins (31). Similarly, long circulating dextran coated iron oxide particles have been shown to be internalized by dividing tumor cells (32). In contrast, standard gadolinium based agents solely image areas of tumor where blood brain barrier breakdown has occurred, often failing to enhance the infiltrating tumor margins. Therefore, areas of tumor that would not been appreciated on gadolinium enhanced MRI can be detected using iron-oxide-based nanoparticles (Figure 2, 33, 34). In addition, unlike gadolinium chelates, which tend to diffuse freely through a tumor and into surrounding brain, iron-oxide based nanoparticles, because of their low diffusivity and the fact that they are phagocytosed by cells within a brain tumor, tend to persist longer within the tumor parenchyma and more accurately delineate tumor margins (31).

Figure 2.

Demonstration of differences between gadolinium enhancing areas of tumor (A) and iron oxide nanoparticle-enhancing areas of tumor (B). Note the area of tumor (solid arrowhead) that enhances with iron-oxide nanoparticles but not gadolinium. (Reprinted from ref 34. Permission requested)

Importantly the nontoxicity of several iron oxide based contrast agents has also been demonstrated. While elemental iron causes the liberation of free radicals in neural tissues, there was no evidence of tissue damage in a recent study on the toxicity of intravenously administered iron oxide nanoparticles. Additionally, when nanoparticle delivery is promoted through transient blood brain barrier relaxation, there is no clear evidence of pathologic changes within the brain (35).

An emerging niche for iron oxide nanoparticle-based contrast agents is intraoperative MRI. One of the greatest challenges in the use of intraoperative MRI is distinguishing enhancing regions of residual tumor from areas where the blood brain barrier has been surgically disrupted. Due to the relatively short half-life of gadolinium, and the relatively lengthy nature of complex tumor resections, a dose of gadolinium must be given intraoperatively to detect residual tumor. Intraoperatively administered gadolinium may cross areas of surgically induced blood brain barrier breakdown resulting in enhancement of tumor free regions in the surgical corridor. Since iron oxide nanoparticles cause enhancement in brain tumors after being phagocytosed by cells within a brain tumor over a much longer period of time (24–48 hours), a single dose of contrast, given 24 hours prior to surgery is sufficient to serve as an intraoperative contrast agent and dose not cause enhancement of surgically manipulated tissues (36). Therefore, in patients receiving iron oxide nanoparticles as a contrast agent, the surgeon can proceed with confidence in removing all enhancing tissue detected with intraoperative MRI.

Iron oxide nanoparticles also hold promise in detecting a therapeutic response within a brain tumor to a given treatment. Anecdotal evidence from a recent pilot study comparing Ferumoxytol, an iron oxide nanoparticle, and gadolinium for brain tumor imaging suggested differential contrast enhancement in previously irradiated tumors compared to recurrent tumors. The same study determined that ferumoxytol, which leaks into a tumor more slowly than gadolinium, provides more accurate measurements of tumor perfusion. Accurate perfusion measurements hold promise in detecting vascular degradation, which may occur in treated tumors prior to shrinkage. Ferumoxytol-enhanced MRI may ultimately serve to be a useful agent in determining the early response of a brain tumor to a given therapy by assessing vascular integrity (30).

Targeted Nanoparticle-Based MRI Contrast Agents

The capacity for highly selective tumor targeting is the second major advantage of iron oxide based nanoparticles over gadolinium. Currently in preclinical development, a number of iron oxide nanoparticles targeted with peptides show promise as highly tumor-specific contrast agents. The first targeted nanoparticle created for brain tumor imaging consisted of dextran coated monocrystalline iron oxide nanoparticles (MION) coated with conjugated to tumor-specific antibodies. This novel agent was shown to be effective in enhancing implanted LX-1 brain tumors in mice (37). Although there has been little follow-up on the nanoparticles described in this report, it demonstrated the concept that nanoparticle-based contrast agents can be specifically targeted to individual tumor antigens in vivo.

An alternative method for targeting nanoparticle-based contrast agents to specific brain tumor antigens relies on coating the nanoparticle surface with drugs. Methotrexate-coated nanoparticles bind to tumor cells and are internalized through interactions with folic acid receptors, which are upregulated on the surface of a variety of cancerous cells, including glioma. Methotrexate can be conjugated to the surface of iron oxide-loaded nanoparticles in a variety of ways to create a multivalently targeted nanoparticle. Preliminary characterization of this type of nanoparticle shows promise for future development as an MRI contrast agent in in vitro studies (38,27). Interestingly, methotrexate-targeted, iron oxide nanoparticles can also be used as a means of delivering methotrexate to cells that can be monitored by MRI.

A third method for targeting nanoparticle-based MRI contrast agents to tumors is through the attachment of peptides such as chlorotoxin and F3 to the nanoparticle surface. Although in vitro characterization of chlorotoxin-targeted iron oxide nanoparticles suggests they will serve as efficient in vivo contrast agents (24), there have been no published reports of their performance in vivo. The first published report of a tumor specific, peptide-targeted, nanoparticle-based contrast agent recently demonstrated the feasibility of targeting a specific subset of cells within a tumor to achieve high quality contrast enhacement (13). The nanoparticle characterized in this report consists of a polyacrylamide gel containing iron oxide crystals and a photosensitizer, (figure 1) targeted with multiple F3 peptides attached to its surface. When injected intravenously F3-targeted, iron oxide-containing nanoparticles provided persistent contrast enhancement of implanted tumors that was more profound and persistent than identical non-targeted nanoparticles (13). In addition to suggesting the feasibility of targeting tumor cells with nanoparticles, this study also demonstrated that peptide targeting promotes the retention or binding of nanoparticles to target cells.

Intraoperative Nanoparticle-Enabled Brain Tumor Delineation

Whenever possible, surgery is the primary mode of treatment for adult and pediatric tumors (39, 40). One of the greatest challenges of brain tumor surgery is achieving a complete resection without damaging crucial structures near the tumor bed. Achieving gross total resection improves survival by lowering the risk of recurrence and reducing tumor cell burden to levels that can be eradicated or controlled with adjuvant therapy (41). In the most common types of brain tumors the ability to achieve radiographically-confirmed gross total resection has been shown to be a key factor in the duration of progression-free survival and outcome (42–47).

Achieving gross total resection currently relies on the neurosurgeon’s ability to judge the presence residual tumor at the time of surgery. Unfortunately, neoplastic tissue that is easily detected radiographically, is often virtually indistinguishable from normal brain. Several studies evaluating the extent of surgical resection highlight the fact that in many cases, especially in diffusely invasive brain tumors, a significant amount of residual tumor remains even after gross total resection (48). Residual tumor left behind after gross total resection may make additional surgery necessary or result in adjuvant treatment failure.

While stereotactic navigation and intraoperative MRI have been utilized to improve extent of resection, their impact on outcome is still controversial (49, 50). The use of fluorescent and visible dyes has long been proposed as a means of visualizing tumor margins intraoperatively (51–53). Preclinical studies on fluorescent dyes suggest that they may improve the extent of resection (54). While the need for a dye based method for tumor delineation has long been recognized, investigators have been hampered by three main difficulties: (1) the inability to stain tumor cells specifically, (2) the inability to concentrate an adequate quantity of dye within tumor cells to achieve visual contrast and (3) the inability to create a dye that would be useful for a wide range of tumors. Dye-loaded nanoparticles have recently been reported to meet each of these challenges.

Two groups have designed iron oxide based nanoparticles loaded with the near-infrared fluorescing (NIRF) molecule Cy5.5 (55, 24). These nanoparticles can be visualized in vitro in 9L cells and in vivo in implanted 9L brain tumors with the use of a specialized fluorescence detection system. Moreover, the particles are detectable on intraoperative MRI because of their iron oxide core. Follow-up studies on Cy5.5-loaded nanoparticles elegantly demonstrate that Cy5.5 loaded iron oxide nanoparticles delineate margins of multiple types of implanted tumors with high accuracy through uptake by tumor cells and surrounding microglia (Figure 3, 56). To improve tumor-specific binding, one group targeted Cy5.5-based iron oxide nanoparticles with chlorotoxin to promote tumor-specific binding and internalization and were tested in vitro (24). Determining the utility of chlorotoxin targeting of Cy5.5-loaded nanoparticles will require in vivo investigation. In addition, one potential limitation of Cy5.5-tagged nanoparticles is that sophisticated fluorescent imaging technology would be necessary to visualize Cy5.5. Furthermore, traditional microsurgical dissection and Cy5.5 visualization could not take place simultaneously as the illumination necessary to visualize the fluorescent Cy5.5 profoundly darkens the operative field.

Figure 3.

Fluorescent staining of implanted GFP-expressing 9L gliosarcoma by nanoparticle containing Cy5.5 (near infrared dye). The tumor is visualized under normal lighting conditions (A), GFP channel (B) and Cy5.5 channel. (Reprinted from Ref 55. Permission requested)

Optical semiconductor nanocrystals called quantum dots are an alternative to Cy5.5-loaded nanoparticles that have been preliminarily characterized for use in the visible delineation of implanted experimental brain tumors. Quantum dots contain a core of hundreds to thousand of group II and VI (e.g. CdSe and CdTe) or group III and IV elements (e.g. InP and InAs), which can be modified to produce emission spectra at specific wavelengths in the 400–2000nm range. The core of a quatum dots is typically coated with an organic polymer and, like other nanoparticles, can be modified with surface molecules that confer specific pharmacokinetic and targeting properties (57). QDot ITK Amino/PEG 705-nm emission quantum dots were recently demonstrated to induce fluorescent staining of implanted C6 brain tumors (58).

Due to their heavy metal content, one drawback to the use of quantum dots is their potential toxicity in normal tissues. Moreover, like Cy5.5-loaded nanoparticles, the quantum dot preparation utilized by Jackson et al. also requires a system for fluorescence imaging to visualize stained tissue and requires a darkened operative field. Still, it is possible that the elemental composition of the quantum dot core could be modified to emit light in the visible (58). Finally, while the quantum dot preparation reported by Jackson et al. was non-targeted, monoclonal antibody targeted fluorescent quantum dots were shown to be superior to non-targeted quantum dots in staining implanted tumor xenografts in mice (59). A recent report also described efficient in vitro targeting of fluorescent quantum dots coated with a monoclonal antibody against epidermal growth factor receptor to the SKMG-3 human glioma cell line (60).

We have recently developed and characterized Nanocyan, a methylene blue-loaded polyacrylamide nanoparticle, that shows promise for clearly delineating neoplastic tissue, without the assistance of additional equipment, under the normal lighting conditions of an operating room (61, 62). Nanocyan consists of a polyacrylamide core and contains a high quantity of methylene blue in its core. The surface of Nanocyan is coated with numerous F3 targeting peptides and a coating of cysteine to prevent non-specific binding. In in vitro studies, Nanocyan has been shown to deeply stain 9L gliosarcoma cells in an F3-dependent manner (see figure 4). These studies represent the first published report of a nanoparticle applied to staining tissue under normal lighting conditions. Importantly, the components of Nanocyan are nontoxic: the polyacrylamide matrix is currently under consideration for FDA approval while methylene blue has been approved for years for human administration. The relevance of Nanocyan for future clinical use will be evaluated as in vivo data are generated.

Figure 4.

In vitro staining of 9L gliosarcoma cells with Nanocyan. Pictures, under normal ambient lighting conditions, of 4×10⁶ washed, pelleted 9L gliosarcoma cells treated with blank polyacrylamide NP (A), non-targeted Nanocyan (B), F3-targeted Nanocyan (C), and HIV TAT-targeted Nanocyan (D) for 4 hours at nanoparticle concentration of 5mg/ml.

Nanoparticle-Based Chemotherapy

As knowledge of the molecular pathogenesis of brain tumors develops, a growing number of targets for novel chemotherapeutic agents are identified (63). One of the greatest challenges in translating novel chemotherapeutics into the clinical realm is developing an efficient means of therapeutic delivery (64). Among the obstacles inherent in delivering chemotherapy to brain tumors is the selective permeability of the blood-brain barrier (BBB) and blood-tumor barrier (BTB). By assisting in passage across the blood brain barrier, nanoparticles can serve as vehicles to enable the efficient delivery of drugs that would otherwise be unable to be delivered at therapeutic levels to brain tumors. Nanoparticles may also serve as a potential solution to the problems of drug-resistance and efficient delivery hydrophobic drugs to tumor cells. Nanoparticles developed as drug delivery devices can be broadly categorized into lipid-based compounds and non-lipid based compounds.

Lipid-Based Drug Delivery Nanoparticles

One of the best characterized lipid-based nanoscale compounds developed for brain tumor drug delivery are, solid lipid nanoparticles (SLN). These nanoparticles are prepared by high-pressure homogenization or micro-emulsion of solid physiologic lipids (65). Although the exact mechanism by which SLN’s cross the BBB and BTB is unknown, internalization is hypothesized to be mediated by endocytosis of SLN’s by endothelial cells. The process of endocytosis is thought to be facilitated by the adsorption of circulating plasma proteins to the SLN surface (66). The lipid matrix of SLN provides a means of loading drugs and protecting them from degradation. The unloading of drugs within target tumor tissues can also be controlled depending on the surface coating of the SLN and its constituent lipids (67). In addition, since they are composed of physiologic lipids, SLN’s are non-toxic (68).

Doxorubicin and paclitaxel are among the extensive list of both hydrophobic and hydrophilic drugs that have been encapsulated within SLN’s (67). Drug-loaded SLN’s have recently been shown to dramatically enhance tumor concentrations and decrease plasma concentrations of doxorubicin and paclitaxel compared to equivalent doses of free drug (64). Proponents of SLN’s speculate that by achieving high drug bioavailability to tumor tissue, SLN’s minimize the therapeutic dose of the drug and, consequently, its systemic toxicity. Further characterization of SLN’s in vitro and in vivo in model systems will be required before their clinical use in brain tumor patients is explored.

Similar in composition to SLN’s, nanoparticle formulations of low density lipoproteins (LDL) have also been proposed as novel drug delivery devices that maybe useful in the future treatment of glioblastoma multiforme. LDL nanoparticles are preferentially taken up by tumor cells, which commonly overexpress LDL receptors due to increased metabolic demand for lipids. The utility of LDL nanoparticles as drug delivery devices has been suggested by in vitro studies that demonstrate their rapid internalization by glioma cell lines via a LDL-dependent mechanism. These studies also identified differences in LDL nanoparticle internalization between established glioma cell lines, most likely due to differences in LDL receptor expression (69).

Ranging from 80nm to 100 μM in size (68), not all liposomes can be strictly defined as nanoparticles. Nonetheless, they bear mentioning in this discussion since several recently developed nanoscale liposomes have promise for the future treatment of brain tumors. Liposomes are spheres created by phospholipid bilayers that can encapsulate both hydrophobic and hydrophilic drugs. Liposomal doxorubicin and daunorubicin were the first liposomal agents applied to the treatment of brain tumors but their efficacy has been disappointing (70, 71), likely as a consequence of poor intratumoral distribution.

However, convection enhanced delivery methods have given new life to the concept of drug-loaded liposomes in the treatment of brain tumors (72). Topotecan is an inhibitor of topoisomerase I induces DNA damage in glioma cells and but has not played a major role in the treatment of brain tumors due to its high systemic toxicity. Recently liposomal topotecan has been prepared which, when administered via convection enhanced delivery to animals bearing implanted U87MG gliomas, has been shown to improve survival greater than 20-fold (73, 74). Liposomal topotecan enables the delivery and accumulation high drug concentration within the tumor while minimizing the systemic drug concentration (74). The tumoricidal effect of liposomal topotecan is thought to stem from its anti-angiogenic properties (73).

The efficiency of liposomal drug delivery devices may be further enhanced through molecular targeting. IL-13Rα2, an interlukin-13 receptor, is upregulated on the surface of glioblastoma cells but not on surrounding cells. This difference has been exploited in the targeting of liposomal doxorubicin. Doxorubicin-loaded nanoliposomes, targeted with conjugated IL-13, were shown promotes internalization and, subsequently, cytotoxicity in U251 glioma cells. Additionally, in vivo studies demonstrated inhibition of the growth of subcutaneously implanted gliomas with weekly nanoliposome administration (25). Since traversing the BBB and intratumoral dispersion were the original stumbling blocks for the clinical application of liposomal doxorubicin, IL-13-targeted lipsomal doxorubicin must be tested in intracranial models of glioma to determine its true therapeutic potential.

Non-Lipid Based Drug Delivery Nanoparticles

Non-lipid based drug delivery nanoparticles are colloidal particles consisting of matrices of synthetic biocompatible, and often biodegradable, polymers. These particles are desirable for drug delivery because they can be chemically modified to adsorb, trap, or bond to drugs. Like lipid-based nanoparticles, non-lipid based nanoparticles isolate their payload from the physiologic environment to prevent degradation and toxicity and can be modified to alter the kinetics of drug efflux in target tissues (68).

Doxorubicin has been shown to have high cytotoxicity in glioma cell lines but its clinical use is limited by the fact that it does not cross the BBB since it is a substrate of p-glycoprotein, a drug efflux system. In addition to the liposomes described above, a number of non-lipid based drug delivery nanoparticles have been developed to enable efficient delivery of doxorubicin to brain tumors in vivo. When rats bearing 101/8 gliomas were treated with doxorubicin-loaded, polysorbate coated nanoparticles, survival time was greatly prolonged compared to free doxorubicin. In addition 20% of rats treated with doxorubicin-loaded nanoparticles remained in long-term remission (75). A follow up study demonstrated that the cytotoxicity of doxorubicin-loaded poly(butyl cyanoacrylate) nanoparticles was greater than that of free drug (76). This observation suggests that the nanoparticle formulation enables the accumulation of a high concentration of drug within tumor cells, possibly by inhibiting the p-glycoprotein drug efflux system (77).

Paclitaxel is a potent chemotherapeutic agent that interferes with microtuble function in tumor cells. While paclitaxel has been shown to have high potent in vitro activity against glioma cell lines (78) it has had little clinical success, due its rapid exclusion from cells via the activity of p-glycoprotein. Moreover, because paclitaxel is a hydrophobic compound, it must be administered with the surfactant Cremophor EL, which frequently causes serious allergic reactions among other side effects (79). Nanoparticle formulations of paclitaxel were developed to improve delivery to glioma cells and to circumvent the need for cremophor EL. Initial paclitaxel-loaded cetyl alcohol/polysorbate nanoparticles were found to be as toxic to tumor cells in vitro as free paclitaxel (80). However more recent paclitaxel-loaded poly(d, l-lactide-co-glycolide) (PLGA) nanoparticles were found to be much more cytotoxic to C6 glioma cells in vitro. This difference is attributed to internalization and intracellular unloading of water-soluble PLGA nanoparticles (81). Although extensive in vivo studies will be necessary to evaluate the chemotherapeutic potential of paclitaxel-loaded PLGA nanoparticles, it is likely that the nanoparticle platform will eliminate the need for Cremophor EL when administering paclitaxel. Data from completed and ongoing clinical trials of Abraxane, a nanoparticle formulation of paclitaxel, in patients with breast cancer, non-small cell lung cancer and melanoma will be useful in determining how nanoparticle-based drugs may be applied to glioma patients (82).

Like paclitaxel, the topoisomerase inhibitor, camptothecin, holds promise as a chemotherapeutic agent but has poor bioavailability because it is minimally soluble in water. A novel non-lipid nanoparticle, consisting of drug-containing micelles coated with magnesium-aluminum layered double hydroxides (LDH), has recently shown promise as a biocompatible delivery device for camptothecin. Camptothecin-LDH nanoparticle treated 9L gliosarcoma cells were killed as efficiently as free camptothecin-treated cells (83). While there was no difference between the in vitro efficacy of camptothecin-LDH nanoparticle and free camptothecin, in vivo the improved stability of the nanoparticle complex would be expected to prevent degradation and improve its therapeutic potential.

Non-Conventional Nanoparticle-Based Anti-Cancer Therapies

Even with the best available surgical, chemotherapeutic and radiation treatments patients with high grade glioma have a poor prognosis (84). Moreover, today’s state of the art treatments are often poorly tolerated and dramatically impact the quality of life in the glioma patient (85). Therefore, novel treatments such as gene therapy, photodynamic therapy and thermotherapy have been proposed. As discussed above, because of their favorable physicochemical properties, nanoparticles have begun to impact the delivery of traditional chemotherapeutic agents. For similar reasons, nanoparticles may also play a future role in the delivery of novel therapies in glioma.

Nanoparticle-Enabled Gene Therapy

Gene therapy is based on the concept that specific exogenous genes can be incorporated into the tumor cell genome produce to produce a tumoricidal effect. Gene therapy is felt to hold promise in the future treatment of glioblastoma multiforme (86). While, viral vectors have traditionally been the primary agents used to deliver genes to target cells they carry the risk of serious immune and inflammatory responses in the host (87). Liposomes were among the first non-viral vectors developed for delivery of gene therapy but were initially limited by low transduction efficiency (88). Their efficiency was increased somewhat through surface ligand targeting via monoclonal antibodies to the transferrin receptor (89). However due to the limitations of liposomes discussed above, nanoparticles have recently been developed as non-viral vectors for gene therapy.

A biopolymeric gene delivery nanoparticle has recently been shown to be effective in vivo in delaying tumor growth. A plasmid encoding proapoptotic Apo2 ligand/tumor necrosis factor-related apoptosis-inducing ligand (Apo2L/TRAIL) was incorporated into a cationic albumin-conjugated pegylated nanoparticle and administered to C6 cells and found to be efficient at incorporating plasmid DNA into the host cell genome. When these nanoparticles were injected intravenously, they were found to accumulate in tumor cells, and, compared with blank nanoparticle, they inhibited tumor growth and prolonged survival in mice bearing implanted C6 gliomas (90).

Nanoparticle-Enabled Thermotherapy

The induction of hyperthermia as an anti-cancer therapy has been proposed since the 1970’s (91) yet its incorporation into the mainstream treatment of cancer has been limited by the difficulty of safely delivering heat deep within bodily structures. Moreover previous attempts at local thermotherapy for the treatment of brain tumors using highly focused ultrasound were hampered by the electromagnetic properties of the skull and the challenge of achieving even temperature distribution throughout a lesion (92). Therefore an alternative strategy was developed in which superparamagnetic iron oxide nanoparticles were injected intratumorally and heated via an alternating magnetic field. This strategy was first applied to implanted rat RG-2 intracranial tumors and shown to be effective in improving survival. The improvement in survival was felt to result from a decreased proliferation of heated tumor cells and, consequently, reduced tumor growth (93). The observations of Jordan et al. corroborated previous work using a similar strategy for delivering thermotherapy to treat subcutaneously implanted gliomas (94).

Data from animal studies prompted a clinical trial of magnetic nanoparticle enabled-thermotherapy in 14 patients with unresectable or recurrent tumors. Although the effects of thermotherapy on the patients in the study were not compelling, the therapy was not associated with any adverse events or toxicity and was deemed to be safe for further investigation in patients with glioblastoma multiforme. A phase II study evaluating nanoparticle-enabled thermotherapy on 65 patients with recurrent glioblastoma multiforme is forthcoming (92).

Nanoparticle-Enabled Photodynamic Therapy

Photodynamic therapy (PDT) is an alternative to current adjuvant therapy that carries little local or systemic treatment associated morbidity and is not susceptible to the development of resistance. PDT relies on activation of a photosensitizer which, when activated by a specific wavelength of light, induces the release of energy to tissue oxygen to generate reactive oxygen species which, in turn, induce cellular toxicity (95). PDT was initially applied clinically to cutaneous and bladder malignancies which can easily be exposed to light due to their location. While brain tumors cannot be exposed to light as easily as superficial tumors, even the deepest brain tumors can be easily illuminated after traditional surgical exposure.

The idea of applying PDT to brain tumors was first suggested by studies in 1972 which demonstrated that PDT mediate cytoxicity in gliomas in vitro and in vivo (96). Since the initial demonstration of PDT efficacy in gliomas, investigators have continued to use hematoporphyrin derivatives as photosensitizers. While the safety of delivering PDT to patients with gliomas has been established (97), the efficacy of PDT for treating brain tumors has been limited in clinical trials, probably because of the difficulty of creating tumor-specific, sufficient accumulation of photosensitizer within neoplastic cells (98). Polymeric nanoparticles offer a solution to this problem by allowing the delivery of a high quantity of photosensitizers to tumor cells via tumor specific ligands. In addition, as discussed above, when encapsulated, photosensitizers, such as methylene blue are protected from degradation by plasma enzymes (16). A targeted nanoparticle containing methylene blue that may be used both as an optical contrast agent and photosensitizer is currently under development in our laboratories.

The concept of nanoparticle-enabled PDT was suggested when McCarthy demonstrated the eradication of subcutaneous implanted U587 gliomas in through PDT mediated by meso-tetraphenylporpholactol loaded poly(lactic-co-glycolic acid) nanoparticles (99). In a study more relevant to the future treatment of human brain tumors, Reddy et al induced long-term remission of implanted 9L gliomas through PDT mediated by F3 targeted, photofrin loaded magnetic nanoparticles. Animals receiving control treatment had a mean survival of 8.5 days while those receiving PDT mediated by targeted nanoparticles had a mean survival of 33 days with a survival of greater than 6 months in two of five rats (13). While these results are compelling, future work will be necessary to evaluate the properties of targeted nanoparticles in biologic systems and their capacity to mediate PDT prior to proposing clinical trials.

Future Challenges

In addition to the traditional difficulties of brain tumor drug delivery, which which have been briefly touched on above, challenges related to nanoparticle pharmacology must be overcome before nanoparticles become a part of the management of brain tumors. When administered systemically, nanoparticles are cleared swiftly by the reticuloendothelial system. This process involves opsonization of nanoparticles, phagocytosis by macrophages and uptake in the liver and spleen (100). Clearance of nanoparticles by the reticuloendothelial system can be partially blocked by the attachment of hydrophilic molecules to their surface (82). However, common agents employed to achieve a hydrophilic coating such as polyethylene glycol or Pluronic can be immunogenic or proinflammatory (60). Moreover, the clearance of polyethylene glycolated nanoparticles is via the liver where they are excreted into the bile and feces. The elimination of nanoparticles through bile is relatively slow, thereby increasing the risk of long-term toxicity of retained nanoparticle remnants. Nanoparticle elimination may be facilitated by the use of nontoxic biodegradable components that can be broken down into subparticles less than 5nm in size that can be excreted renally (101).

Because the field of nanotechnology is relatively young, the long-term health effects of nanoparticles are unknown. One approach to estimating the possible toxicity of nanoparticles is to examine the health effects of nanoparticle-sized airborne products called ultrafine particles that are commonly released as manmade and naturally occurring pollutants. Ultrafine particles are known to have detrimental effects on the cardiovascular system by alterning cardiac autonomic function and promoting inflammatory changes that are involved in atherosclerosis and cardiopulmonary embolic events (8). While the relevance of ultrafine particle toxicity to injected nanoparticles designed for human use is unclear, extensive studies on the pharmacokinetics and toxicity of nanoparticles will be necessary before they are implemented for use in human disease.

Summary

Nanoparticles have the potential for advancing the diagnosis, operative management and adjuvant therapy brain tumors in the future. Nanoparticle-based MR contrast agents have the potential to uncover portions of tumor, especially along the tumor-brain interface, that would have been unclear with traditional gadolinium-enhanced MRI. In addition, several nanoparticles, developed with the objective of allowing the surgeon to see tumor margins intraoperatively, may ultimately enable neurosurgeons to improve the extent of brain tumor resection. Moreover, the delivery of chemotherapy and non-traditional therapies to brain tumors will likely be greatly improved by nanoparticle based-drug delivery devices. As our knowledge of their pharmacology and long-term health effects expands, nanoparticles are likely to play a central role in the future management of the brain tumor patient.