Image-guided interventional therapy for cancer with radiotherapeutic nanoparticles

Abstract

One of the major limitations of current cancer therapy is the inability to deliver tumoricidal agents throughout the entire tumor mass using traditional intravenous administration. Nanoparticles carrying beta-emitting therapeutic radionuclides that are delivered using advanced image-guidance have significant potential to improve solid tumor therapy. The use of image-guidance in combination with nanoparticle carriers can improve the delivery of localized radiation to tumors. Nanoparticles labeled with certain beta-emitting radionuclides are intrinsically theranostic agents that can provide information regarding distribution and regional dosimetry within the tumor and the body. Image-guided thermal therapy results in increased uptake of intravenous nanoparticles within tumors, improving therapy. In addition, nanoparticles are ideal carriers for direct intratumoral infusion of beta-emitting radionuclides by convection enhanced delivery, permitting the delivery of localized therapeutic radiation without the requirement of the radionuclide exiting from the nanoparticle. With this approach, very high doses of radiation can be delivered to solid tumors while sparing normal organs. Recent technological developments in image-guidance, convection enhanced delivery and newly developed nanoparticles carrying beta-emitting radionuclides will be reviewed. Examples will be shown describing how this new approach has promise for the treatment of brain, head and neck, and other types of solid tumors.

1. Introduction

1.1. Challenges in drug targeting and delivery to solid tumors of intravenously administered drugs

One of the major challenges of current cancer therapy is the inability to deliver intravenously administered tumoricidal drugs throughout the solid tumor mass. One reason for this is that intravenously administered drugs are inhibited in their intratumoral penetration by high interstitial pressures which prevent diffusion of drugs from the blood circulation into the tumor tissue [1–5]. This problem is compounded by the relatively rapid clearance of intravenously administered drugs from the blood circulation by kidneys and liver. In addition, drugs that do reach the solid tumor by diffusion are inhomogeneously distributed at the micro-scale. This problem of inadequate intratumoral drug levels cannot be overcome by simply administering larger systemic doses as toxicity to normal organs is generally the dose limiting factor. The use of nanoparticles for carrying anti-cancer drugs is one method for increasing the drug accumulation in tumor following intravenous administration since the nanoparticles can be passively targeted and accumulate in the tumor through the enhanced permeability and retention (EPR) effect [6–8]. However, even nanoparticulate drugs have poor penetration from the vascular compartment into the tumor and the nanoparticles that do penetrate are most often heterogeneously distributed [9–11]. Imaging methods at the micro-scale are being developed to better understand the heterogeneous pattern of nanoparticle accumulation in an attempt to develop new therapies [12–14].

1.2. Inclusion of imaging in drug delivery studies

Imaging is becoming an integral component of drug development as well as for monitoring drug delivery and the response of targeted processes to the therapy [15–17]. Imaging can be used to guide minimally invasive procedures such as guiding a needle for tumor biopsy which is much less invasive than collecting specific tumor samples surgically [18]. Companion imaging probes targeting molecular features determined from the biopsy sample can be integrated into the drug development process. In addition, the inclusion of a companion imaging probe during drug development can aid in determining the clearance kinetics and tissue distribution of the drug non-invasively using imaging modalities such as single photon emission computed tomography (SPECT), positron emission tomography (PET), X-ray computed tomography (CT), magnetic resonance imaging (MRI), ultrasound or optical methods [19]. This companion imaging probe can also be used to determine the likelihood of the drug reaching the tumor and to what extent. In this situation of personalized medicine, individual cancer patients can be stratified for promising drug treatment responses with this type of imaging. Drugs that have increased accumulation within the targeted site are likely to be more effective as compared with others with minimal accumulation at the target site [19]. This makes treatment more efficient and cost effective. Moreover, the Food and Drug Administration requires the availability of a companion diagnostic test to select patients for targeted therapies and in many cases this diagnostic is an imaging agent [20,21].

Nanoparticle-based drugs have an additional advantage over free drugs with their potential to be multifunctional carriers capable of carrying both therapeutic and diagnostic imaging probes (theranostic) in the same nanocarrier. These multifunctional nanoparticles can serve as theranostic agents and facilitate personalized treatment planning. Additionally, nanoparticles are less likely to be affected by inclusion of an imaging component within their structure unlike small molecules, peptides, oligonucleotides and proteins (monoclonal antibodies) which can more readily lose functionality by the addition of imaging probes. The design and testing of potential theranostic nanoparticles has been a burgeoning area of research in the past 15 years. An exhaustive review of these nanoparticle constructs is outside of the scope of this article and previous review articles covering this topic are available in the literature [16,22–29].

1.3. Image-guidance for enhancing drug delivery of intravenously administered nanoparticle-based drugs

Imaging can also be used for localization of the tumor to improve the placement of a catheter or external device within tumors to cause cell death through thermal ablation or oxidative stress secondary to reactive oxygen species. Image guided thermal ablation methods include radiofrequency (RF) ablation, microwave ablation or high intensity focused ultrasound (HIFU) [30–33]. Photodynamic therapy methods using external light devices to activate photosensitizing agents can also be used to treat superficial tumors or deeper tumors when used with endoscopic catheters.

A number of nanoparticle types including iron oxide particles, carbon nanotubes, gold nanoparticles, polymeric nanoparticles, and liposomes have been tested as combination therapies to improve the efficiency of the ablation through mechanisms such as increased tumor heating and enhanced cell death in the presence of nanoparticles as well as triggered release of anti-cancer agents from the nanoparticle during the ablative procedure. A recent review of nanoparticles applied to thermal ablation has been published by Manthe et al. [34].

1.4. Radiotherapeutic nanoparticles

Over the past 20 years, wide ranging investigations have been performed into the use of nanoparticles for imaging and therapy applications, including those with liposomes, solid lipid nanocapsules, polymeric nanoparticles, dendrimers, micelles, gold nanoparticles, quantum dots, iron oxide magnetic particles, and carbon nanotubes. Several review articles have been published describing these nanoparticles [35–46]. Despite the numerous investigations of nanoparticles for cancer therapy, there has been limited effort in the development of radiotherapeutic nanoparticles for image-guided radionuclide therapies. To date, the majority of investigations of nanoparticles with radiotherapeutic nanoparticles have focused on beta (β)-emitting radionuclides. This is most likely due to their much greater penetration range in tissue (1–4 mm in average tissue penetration), especially in comparison with alpha (α)-emitters (10–100 µm in average tissue penetration). Table 1 describes the β-emitting radionuclides that have properties suitable for use in image-guided nanoparticle radionuclide therapy with the corresponding nanoparticles that have been prepared. When considering techniques for radiolabeling nanoparticles, cost, convenience and the ability to prepare radiolabeled nanoparticles with high radiolabeling efficiency and radiolabel stability are important factors. Stability of the radionuclide within the nanoparticle is very important. The radionuclide needs to stay associated with the nanoparticle for a minimal period of several days in order for adequate therapy to be delivered to the tumor otherwise the advantageous distribution characteristics of the nanoparticle will be lost. For historical purposes, Table 1 also includes phosphorous-32 (P-32) colloidal particles, the majority of which are in the micro-particle size range.

Table 1.

Properties of β-emitting radionuclides and their use in radiotherapeutic nanoparticle preparation.

Radionuclide		Average β energy (MeV)	Mean path length (mm)	Half-life (days)	Gamma energy (keV) (% decay)	Production method	Nanoparticle radiolabeled
Copper-67	67Cu	0.141	0.71	2.6	185 (48.7%)	Reactor	No 67Cu. Gold nanoshells have been labeled with diagnostic PET imaging 64Cu [47]
Gold-198	198Au	0.315	1.6	2.7	412 (95.6%)	Reactor	Gold dendrimer nanodevice [48]
							Gold nanoparticle [49,50]
Holmium-166	166Ho	0.666	3.2	1.1	81 (6.7%);	Reactor	Poly-l-lactide nanoparticles [51]
							Liposomes [52]
Iodine-131	131I	0.182	0.91	8.0	365 (82%)	Reactor	Boronated dendrimer [53]
Lutetium-177	177Lu	0.133	0.67	6.7	113 (6.4%)	Reactor	Metallofullerene [54]
							Poly(dl-lactide-co-glycolide) nanoparticles [55]
							Liposomes [56]
Phosphorus-32	32P	0.695	2.9	14	N.A.	Reactor	Chromic phosphate colloid [57–61]
Rhenium-186	186Re	0.362	1.8	3.8	137 (9.8%)	Reactor	Colloid [62]
							Liposomes [38,63–66]
Rhenium-188	188Re	0.764	3.5	0.71	155 (15.6%)	Tungsten-188/rhenium-188 generator	Solid lipid nanocapsules [67–69] Liposomes [70–73]
							Dendrimer [74]
							Iron oxide [75]
							Albumin magnetic nanoparticle [76]
							Quantum dot [77]
							Micelle [78]
Yttrium-90	90Y	0.935	3.9	2.7	N.A.	Reactor	None

1.5. Ideal theranostic properties of certain beta-emitting radionuclides

As noted in Table 1, many beta-emitting radiotherapeutic nuclides used to label nanoparticles also emit a low ratio of gamma photon (~5–15 gamma photons for every 100 beta-emissions). This ratio is considered ideal because it is a high enough ratio of gamma photons for easy monitoring of intratumoral and organ distribution within the body without being so high as to deliver a significant radiation dose to the whole body or to nearby persons. Certain therapeutic radionuclides not only have an ideal ratio of gamma photons, but their photon energy is also in the ideal range for nuclear imaging, ranging from 80 to 200 keV. This energy range is ideal because it is a high enough energy for many of the emitted photons to escape the body, but a low enough energy to be easily stopped by nuclear imaging camera detectors. This is why the most commonly used diagnostic nuclear medicine imaging radionuclide is the pure gamma photon emission radionuclide, technetium-99m (^99mTc), which has an ideal photon energy of 140 keV. The gamma photon emitted by these therapeutic radionuclide containing nanoparticles can be used to localize and quantitate the distribution of the nanoparticles with SPECT nuclear imaging. Therefore, these radionuclides are intrinsically theranostic in nature and this feature has greatly facilitated therapeutic nanoparticle development and these theranostic properties should prove valuable in clinical radiotherapeutic nanoparticle applications. One good example of a therapeutic radionuclide with ideal theranostic properties is rhenium-186 (¹⁸⁶Re). It has ~1 gamma photon for every 10 beta-emissions and these gamma photons have a photopeak energy of 137 keV.

2. Image-guidance for thermal ablation combined with intravenous nanoparticles

2.1. Radiofrequency ablation combined with chemotherapeutic nanoparticles

Image-guidance can be used to direct focal thermal therapy into solid tumors. This thermal ablation can be combined with intravenously administered nanoparticles to increase the effectiveness of the local tumor therapy. For example, image guidance is routinely used to direct a radiofrequency (RF) emitting needle probe into a solid tumor to induce short duration heating of tumor tissue to >50 °C. When long circulating nanoparticles are injected intravenously immediately following this image-guided RF thermal damage, a greatly increased accumulation of the nanoparticles in the region of the thermal ablation occurs. This increased nanoparticle accumulation is most likely due to the increased leakage of nanoparticles into the tumor as well as their increased retention within the region associated with the thermally induced inflammation which has been described as an EPR effect [6,8]. When nanoparticles carry anti-cancer agents, the size of the tumor ablation can be significantly increased as compared to thermal tumor ablation alone [79]. Studies have shown that the size of the tumor ablation zone is much larger when the drug is carried by a nanoparticle as compared to when the drug is free [80].

The potential to induce an increased tumor ablation zone has important therapeutic implications for solid tumor therapy. For example, although RF ablation is widely used in clinical practice, particularly for liver tumor therapy, it is often impossible to completely ablate the entire tumor due to decreased efficacy of this therapy at the highly vascularized margins of the tumors, particularly in tumors that are larger than 3 cm in diameter [81–83]. Although it is possible to achieve local control of liver metastasis in 80–90% of cases with RF ablation if the tumor is less than 2.5 cm in diameter [80], if the tumor is >2.5 cm, local tumor recurrence is common [80]. For this reason, combinations of image-guided thermal ablation with intravenously administered nanoparticles carrying a variety of anti-cancer agents has been investigated as a method to increase the effectiveness of tumor therapy [79,80, 84–86].

Extensive investigations include combinations of thermal ablation and different chemotherapeutic agents in both free form and contained within nanoparticles. In an initial study, Ahmed and Goldberg examined RF ablation combined with different formulations of chemotherapeutic agents [80]. In this study, intravenously administered nanoparticles containing doxorubicin administered at the time of image-guided RF ablation were found to have an increased amount of coagulated tumor tissue and more doxorubicin was deposited within the ablated tumor [80]. In another study by the same group, the tumor destruction diameter for RF ablation alone was 6.7 ± 0.6 mm compared with 11.4 ± 1.1 mm for RF ablation combined with free doxorubicin and 13.1 ± 1.5 mm (P < 0.001) for RF ablation combined with liposomal doxorubicin [79]. The greater tumor destruction found by combining intravenously administered liposomal doxorubicin with thermal ablation therapy was considered to be most likely due to a combination of increased intratumoral drug accumulation and improved cytotoxicity related to thermal effects of the RF ablation [80]. These studies were the first to show the importance of lipid nanoparticles as carriers of the drug when using thermal ablation since the nanoparticle liposome carrier was associated with an increased accumulation of doxorubicin in the tumor while non-encapsulated free doxorubicin did not have increased tumor uptake following RF ablation [80]. In fact, some studies have shown that RF ablation followed by intravenous liposomal doxorubicin was associated with intratumoral doxorubicin uptake that was 5.6-fold greater as compared with unablated tumors in a breast tumor model [87]. In addition to increased tumor cytotoxicity secondary to the increased intratumoral concentrations of doxorubicin, evidence also suggests that the lipid component of liposomes may be contributing to increased tumor damage secondary to formation of lipid hydro-peroxide leading to enhanced oxidative stress [88]. Nanoparticle size can influence the intratumoral drug accumulation and tissue coagulation. Greater tumor coagulation was observed with liposomes of 100 nm diameter compared with either 20 nm or 250 nm liposomes containing doxorubicin [84].

The significantly increased RF ablation-induced uptake of liposomal doxorubicin within tumor tissue was also shown in a head and neck model using imaging with liposomal doxorubicin radiolabeled with ^99mTc or ¹⁸⁶Re [89,90]. These reports used a radiolabeling method for loading a commercial liposomal doxorubicin formulation with high activities of ^99mTc or ¹⁸⁶Re by taking advantage of the same pH gradient used to load doxorubicin into the liposomes [7]. Head et al. demonstrated that radiolabeled liposomal doxorubicin had a 3-fold increased accumulation in tumors treated with RF ablation and that intratumoral doxorubicin levels correlated with the quantitatively measured activity in the tumor using non-invasive nuclear imaging [5].

2.2. Radiofrequency ablation combined with radiotherapeutic nanoparticles

Building on the prior work with combination image-guided RF ablation and liposomal doxorubicin, Soundararajan et al. investigated the addition of radiotherapeutic nanoparticle therapy to the combination of RF ablation and doxorubicin nanoparticles by radiolabeling the liposomal doxorubicin with ¹⁸⁶Re [86,91]. Cancer therapy with these dual therapeutic carrier liposomes has been previously termed chemoradionuclide therapy [90]. When these dual therapeutic nanoparticles are combined with a third therapy of RF ablation, a further increased amount of tumor tissue can be effectively treated [86]. As can be observed in Fig. 1A, tumors treated with triple therapy had increased tumor accumulation of the ¹⁸⁶Re-liposomal doxorubicin which corresponded to greatly decreased tumor growth compared to all single and double therapy combinations (Fig. 1B) and resulted in the highest percentage of complete response to therapy (Fig. 1C). The triple therapy animals also had increased survival as compared to either the RF alone, liposomal doxorubicin alone or various combinations of dual therapies [86]. The mechanism behind the increased response of adding local radiation to the chemotherapy is unknown, but may be due to the synergistic effects of radiation with chemotherapy or due to the local 2 mm beta-particle radiation enhancing the penetration of the drug into the tumor. Although detailed studies of the mechanism need to be performed [85], triple therapy with chemoradiation nanoparticles appears promising for image-guided cancer therapy and clinical trials of this therapy are deserving.

Fig. 1.

(A) Planar gamma camera images of rats bearing head and neck tumor acquired following intravenous infusion of rhenium-186 (Re-186) liposomal doxorubicin (Doxil) without (upper panel) or in combination with RF ablation therapy (lower panel) over a 5 day period. There is increased accumulation of rhenium-186 Doxil in the tumor following RF ablation. Panel B demonstrates the greatly decreased change in tumor volume with the triple therapy of RF ablation (RFA) with liposomes carrying rhenium-186 and doxorubicin as compared with dual and single therapies. Panel C demonstrates the greatly increased percentage of complete response to therapy with triple therapy as compared with dual and single therapy combinations. Complete response was defined as >20% decrease in tumor volume at the end of the study.

A significant advantage of adding ¹⁸⁶Re to this therapy is that ¹⁸⁶Re has an ~10% emission of a gamma photon of 137 keVin the ideal energy range for nuclear imaging. The ability to image the distribution of the ¹⁸⁶Re-liposomes permits quantification of their location within the tumor and within the body. This theranostic ability of ¹⁸⁶Re-liposomal doxorubicin is shown in Fig. 1A in images of rats with head and neck tumors obtained over serial time points out to 5 days. In the rat treated with RF ablation (lower panel),it can be observed that the tumor uptake was significantly higher and there was prolonged retention in the tumor as compared with a rat with an unablated tumor that was also administered the same dose of ¹⁸⁶Re-liposomal doxorubicin. This ability to localize and quantify distribution permits quantification of the dose of liposomes, radionuclide and doxorubicin that accumulates in a specifically treated tumor enabling both chemo and radiation dosimetry.

2.3. High intensity focused ultrasound combined with nanoparticle therapeutics

Another area of significant research is the application of HIFU to improve the drug delivery and triggered drug release from nanoparticles. Several good reviews on the use of HIFU in nanoparticle drug delivery have been published [34,92–94]. Ultrasound technology can impart both thermal and mechanical effects to improve drug delivery from nanocarriers. For mechanical effects, gas bubbles in the mm size range are used to increase the permeability of nanoparticles through the vessel wall or blood brain barrier following intravenous infusion or through sonoporation of cell membranes, thereby enhancing drug and gene transport into the cell [95–106]. Thermal effects from HIFU have been used for tumor ablation and mild hyperthermia. One approach is to use ultrasound image guidance to localize the tumor for HIFU therapy following intravenous infusion of the nanocarrier. The increased heating in the area causes increased permeability of the nanocarrier in the tumor. Various types of nanoparticles are under study for combination therapies with HIFU including polymeric micelles, nanobubbles, nanodroplets, microemulsions and liposomes including specially designed temperature sensitive liposomes [107–113]. To track the drug delivery in many of the studies, the nanocarriers are designed to contain various contrast agents including MRI, fluorescent and radiolabeled agents for non-invasive imaging [114–116]. Increased permeability was shown with ^99mTc-liposomes with more accumulation of the ^99mTc-liposomes in cat fibrosarcomas treated with HIFU hyperthermia [117]. Other groups have also tracked the accumulation of liposomes after HIFU with radiolabeled liposomes [118,119]. Several researchers are combining MRI to monitor drug release following heating of the temperature sensitive liposomes [120,121]. An advantage of MRI is that it can provide chemodosimetry or dose painting feedback to quantify the drug from the signal of MRI contrast within the tumor [92,122,123].

3. Intratumoral administration of nanoparticle drugs

The ability to adequately treat solid tumors at their primary site locally remains an important problem worthy of the development of new therapeutic approaches. Although cancer is generally considered to be a systemic disease, inability to control local cancer invasion remains an important cause of death in ~30% of all cancer patients [124, 125]. It is also likely that inadequately treated local cancer is associated with the development of cancer drug resistance which subsequently leads to locoregional or distant metastasis that go on to result in the death of the patient.

To address the problems of inadequate solid tumor targeting and drug delivery, numerous technologies using direct image-guided intratumoral administration of therapeutic agents are currently undergoing evaluation to avoid the need for drug delivery via systemic administration. Nanoparticles carrying therapeutic radionuclides have recently been shown to have important advantages for use in intratumoral solid tumor therapy. These recent developments using intratumorally administered radiotherapeutic nanoparticles as a viable option for use in cancer patients will be one focus of this review article.

3.1. Convection enhanced drug delivery

To improve drug delivery to solid tumors, an alternative to intravenous drug delivery has been developed known as convection enhanced delivery (CED). CED for cancer therapy is an approach that uses the direct injection of agents into solid tumors [126,127]. CED is performed by using image-guidance to place a catheter tip directly into solid tumor parenchyma while therapeutic agents are infused through the catheter tip under constant pressure over an extended time. The resulting pressure gradient moves therapeutic agents through the interstitial spaces of the targeted tumor by bulk flow, which results in the injected agent being more evenly distributed at higher concentrations over a larger area than is possible without CED [126,128]. By the use of CED, a much higher tumor-to-normal tissue concentration of therapeutic agents is achieved as compared with intravenous administration, which relies solely on the diffusion of drug from the blood into the tumor. Studies have shown that CED also may have important advantages in comparison with simple intratumoral bolus injection [129]. CED was originally described in an animal model in 1994 [130] when Bobo et al. demonstrated that CED enhanced the volume of distribution of a radioactively labeled large molecule in the brain while also achieving concentrations in brain tissue that were orders of magnitude greater than systemic levels [130].

The majority of current investigations using CED have been in the brain due to the limited ability of intravenously administered drugs to cross the blood–brain barrier. Intracerebral CED has been investigated for a variety of disease processes including Parkinson’s and Alzheimer’s diseases in addition to its use to treat brain tumors [131,132]. For brain cancer some of the potential drugs that are being tested by CED include immunotoxins, liposomal agents and gene therapies [22,126,132–139]. CED also has important potential for use in all types of solid tumors, since all solid tumors tend to have high tumor interstitial pressure, limiting drug diffusion from the blood [1]. Therefore, solid tumors of all types are candidates for image-guided nanoparticle therapy, particularly those tumors that are not amenable to surgical removal or when the patient is not a surgical candidate. To date, intratumoral nanoparticle delivery has been investigated with head and neck tumors, prostate and melanoma tumors [48–50,64] in addition to much prior research of CED for brain tumors [128,140–145]. In addition to these solid tumors, it is the author’s opinion that CED delivery with nanoparticle therapeutics also could have applications in lung, liver and breast tumors.

3.2. CED nanoparticle therapeutic as possible adjuvant therapy

CED administration of nanoparticle therapeutics has important promise as an adjuvant for solid tumor therapy. One adjuvant approach is to use CED administered nanoparticles to completely treat the margins of the tumor where residual invasive tumor cells are likely to reside. This can be done by CED administration of nanoparticles along the surgical margins of a tumor cavity during the time of the surgery as proposed by Sampson et al. [129]. CED administered nanoparticles along the surgical margins may have a more homogeneous distribution along the margins of the surgery to treat infiltrative residual tumors than CED administered nanoparticles directly into the tumor [129].

This adjuvant approach has been investigated in a head and neck solid tumor model. In this study, Wang et al. administered ¹⁸⁶Re-radiotherapeutic nanoparticles directly into residual tumor as a model of surgical tumor resection with residual tumor in an intraoperative procedure [146]. In this model, the post-surgical adjuvant therapy was very effective. Control untreated residual tumors had average tumor volumes of 290% of initial volumes at the end of the study (35 days post-treatment) while ¹⁸⁶Re-nanoparticle treated residual tumors had tumor volumes of only ~26% of their initial volumes. All groups showed consistent increases in body weight and there was no significant systemic toxicity observed in any of the animals. With excellent tumor suppression and minimal side-effect profile, the authors concluded that intraoperative use of radiotherapeutic nanoparticles may play a role in the management of positive surgical margins in advanced head and neck squamous cell tumors.

CED nanoparticle therapeutics may also have potential in a neoadju-vant role of treating the central regions of the poorly vascularized tumors prior to administration of systemic therapy. This would have the potential of treating the central portions of the solid tumor that are difficult for intravenous drugs to reach while also potentially increasing intravenous nanoparticle delivery of drugs to the outer portions of the tumor due to EPR effects associated with the CED therapy-induced inflammation.

3.3. CED challenges

Although CED has significant advantages in cancer drug delivery, in initial clinical research it has faced significant challenges related to the 1) inability to verify the distribution and retention of the CED administered drugs, 2) inability to accurately locate the catheter within the tumor, 3) backflow of CED administered drug along the catheter [147], and 4) lack of use of carrier agents, such as nanoparticles, to improve the distribution and retention of therapeutic agents within the tumor. Many of these technical limitations have been discussed in more detail by several authors [134,148]. In the last 10 years, significant progress has been made in all of these areas including 1) significant improvements in image-guidance technologies, 2) improvements in catheter design [132], 3) development of new nanoparticles and 4) development of techniques to improve the intratumoral distribution and retention of CED infused agents based on newly gained scientific understandings of the many factors that influence CED therapy [140]. Another important area where progress has been made is the in vivo imaging technology to monitor the CED infusate distribution. Some of the tracers used in conjunction with the therapeutic agent are radiolabeled albumin, and gadolinium alone or conjugated to albumin or liposomes [135,141, 149–151]. A recent article by Allard et al. provides a thorough review of the current state of our knowledge on the delivery parameters and nanoparticle characteristics for successful convection enhanced delivery of nanoparticles to brain [152]. In recent years, many of the challenges associated with CED therapy have been addressed and these new approaches will be discussed in the following sections of this article.

3.4. Advantages of using β-emitting radionuclides with intratumoral CED nanoparticle therapy

β-Emitting radionuclides provide significant unique advantages for use with nanoparticles in the treatment of cancer. The path length of β-particles in tissue depends on the β-energy of each particular radionuclide [20]. For therapeutic β-emitting radionuclides, the β-particles have path lengths of 1–5 mm in tissue, creating a micro-field of therapy around each nanoparticle in which they are contained (Fig. 2). The 1–5 mm average path length is optimal for delivering high radiation absorbed doses throughout the tumor; while limiting the radiation absorbed dose to a very low level in the surrounding tissue. Considering that the average cell diameter in the human body is ~25–50 µm, the micro-field around a radiolabeled nanoparticle with an average β-emission path length of 2 mm, such as ¹⁸⁶Re, can potentially treat a micro-field of 40–80 tumor cells extending radially in all directions from the nanoparticle. Higher energy β-radionuclides such as yttrium-90 (⁹⁰Y) and rhenium-188 (¹⁸⁸Re), with average β-emission path lengths of 4 mm, can treat a micro-field of 80–160 cells in all directions. This micro-field greatly decreases the need for homogeneity of the actual nanoparticle distribution resulting from CED infusion. As one investigator has stated for effective cancer therapy, β-emitting radionuclides only need to remain within an “effective range” of the tumor cells for a “sufficient period” of time [54]. In contrast, the requirements for nanoparticle distribution homogeneity are much stricter for successful chemotherapeutic drug delivery, in which, generally, the drugs need to escape from the nanoparticle carrier and eventually reach every single tumor cell.

Fig. 2.

Illustration depicting the β-radiation from radiotherapeutic nanoparticles following intratumoral CED administration. A significant advantage of β-emitting radionuclides for use with CED-administered nanoparticles is that the path length of the β-radiation extends the treatment to a micro-field of 40–80 tumor diameter distances from the location of the nanoparticle in all directions.

Because β-emitting radiotherapeutic nanoparticles are well dispersed and retained in solid tumors, they can deliver very high radiation doses to the tumor while greatly sparing adjacent normal tissue. In studies in our lab, glioma tumors in rats were treated with absorbed doses as high as 1700 Gray (Gy) following CED intratumoral administration of ¹⁸⁶Re-liposome nanoparticles without significant toxicity to adjacent normal brain tissue due to the relatively short path length of the β-particle radiation [142]. Initially, 50–65% of the ¹⁸⁶Re-liposome nanoparticles were retained in the brain tumor and only very slowly left the brain over a 5-day time period. The radiation dose delivered to normal organs was also very low because 35–50% of the nanoparticles that initially escaped from the brain tumor were taken up by the liver and spleen which resulted in low doses of radiation being delivered to these organs as determined from serial whole body quantitative scintigraphic imaging. These organs, with weights several hundred times larger than the CED treated tumor, received much lower radiation doses than the tumor because the concentration of any radiation delivered on a per organ weight basis is much lower than the highly focal β-radiation activity concentrated in the tumor.

The 1700 Gy radiation dose that was delivered to the tumor is much higher than can be safely delivered to tumors with external beam radiation where maximum delivered doses are generally no more than 70 Gy due to the limitation of radiation toxicity to the normal organs through which the external beam of radiation passes. Although locally applied β-emitting radiation delivered with nanoparticles can be considered a type of brachytherapy, this type of brachytherapy is very different from current clinical brachytherapy in which metallic seeds carrying low energy photon radiation are implanted in a stationary location within tumors such as prostate cancer [153]. Because photon radiation emitting agents are used with this standard brachytherapy, there is potential for significant radiation toxicity to be delivered in normal organs adjacent to the implanted seeds. Similar to external beam radiation therapy, current clinical brachytherapy use radionuclides with photon radiation path lengths of 1–3 cm which limits the total tumor radiation dose that can be delivered to generally no more than 110–145 Gy [154].

Another inherent advantage of certain β-emitting radionuclides is that they also can act as theranostic agents. Radionuclides such as ¹⁸⁶Re and ¹⁸⁸Re emit gamma photons in a ratio of approximately 10–15% of the number of β-emissions. This relatively low ratio of gamma photons can be used for SPECT imaging to determine the location of the radiation dose delivered to specific regions within the body on a quantitative basis. β-Emitting radionuclides which also emit gamma photons for diagnostic imaging purposes are shown in Table 1. This theranostic capability of specific β-emitting therapeutic radionuclides has proven very useful in the development of rhenium-nanoparticle intratumoral therapies, permitting visualization and quantification of the distribution of these nanoparticles within the tumor and in the various organs of the body on a percent injected dose basis using non-invasive imaging [64,66,91,142,146,155].

3.5. Recent technological progress in image-guidance

Over the last decade, significant improvements have occurred in clinical imaging technology. Magnetic resonance imaging (MRI) scanners with high strength magnetic fields of 3 T or greater have become commonly available in clinical radiology facilities providing physicians with images with much higher resolution and contrast. Localization of tumors based on physiologic processes such as tumor glucose metabolism using positron emission tomography (PET) imaging fused with high resolution X-ray computed tomography (CT) cameras has become available in many hospitals. Recent progress in this clinical diagnostic imaging technology permits ever more accurate localization of pathological processes based on both anatomic structures and physiologic processes. These recent improvements in diagnostic PET/CT imaging cameras and methods has increased the accuracy in the placement of needles and catheters into tumors for more accurate biopsy assessment of tumor masses [156].

PET/CT guided biopsy of a glucose-avid tumor region has increased the accuracy of biopsy sampling and new methods of improving localization by PET/CT are currently under development [157]. The most recent advancement in clinical imaging is the introduction of PET/MRI systems that offer the potential of improved anatomic resolution and tissue characterization with MRI, and the ability to fuse high resolution MRI images with important physiologic processes such as cell division and glucose metabolism using simultaneously acquired PET images [158,159]. In addition to providing high resolution imaging that does not require ionizing radiation, MRI also provides the ability to monitor the intratumoral distribution of CED administered nanoparticles either after treatment or in real time during treatment using imaging nanoparticles that contain MRI contrast agents. PET/MRI image co-registration allows the development of CED infusion protocols to ensure that the most metabolically active regions of a solid tumor received a high concentration of nanoparticles. Nanoparticles are ideally suited for use with these hybrid imaging systems as their versatility allows the same nanoparticle to carry different diagnostic and therapeutic agents within the same nanoparticle and thereby take advantage of the strengths of each type of imaging modality and/or various combinations of different therapeutic agents [114,160]. An example of a multifunctional nanoparticle that can carry therapeutic agents as well as MRI, optical and PET imaging agents for monitoring intratumoral CED delivery of liposomes will be described in a later section of this article.

Building on these image-guidance technological improvements related to biopsy, image-guided cancer therapies have become increasingly popular in clinical practice. Common clinical applications of image-guided cancer therapy in current clinical use include image-guided placement of probes within liver tumor for their thermal ablation [161–163] and the image-guided delivery of chemotherapeutic and β-emitting radiotherapeutic microspheres administered intra-arterially guided by fluoroscopic angiography [164].

4. Advantages of nanoparticles for intratumoral CED administration

Nanoparticles have been found to have significant advantages when used for CED delivery which offer the promising potential to greatly enhance the efficacy of image-guided CED solid tumor therapy. Their principle advantages of enhanced dispersion and retention (EDR) will be reviewed in sections 4.1 and 4.2.

4.1. Nanoparticles have improved retention within the tumor

Specific nanoparticle carriers can have greatly improved retention of therapeutic agents within the tumor following CED administration in comparison with naked, i.e. non-nanoparticle associated, molecules. For brain CED studies, free drugs in solution quickly leave the brain through the blood and cerebrospinal fluid. Much of the research on CED nanoparticle delivery for brain cancer therapy has been in normal brain rather than tumors. Studies by Bankiewicz’s group have demonstrated the very prolonged retention of liposomal drugs following CED administration in a rat U87 tumor model [159]. CED administered topotecan encapsulated within liposomes had a tumor clearance half-life of 1.5 days versus free topotecan which had a tumor clearance half-life of 0.1 days following CED administration [144,149].

In a study published by the authors, the tumor retention and effectiveness of ¹⁸⁶Re-liposomes were compared with a ¹⁸⁶Re-radiolabeled lipophilic small molecule (¹⁸⁶Re-N,N-bis[2-mercapto-ethyl]-N′,N′-diethyl-ethylenediamine (BMEDA)) and free ¹⁸⁶Re-perrhenate after intratumoral infusion in a head and neck tumor xenograft rat model [64]. This study was specifically designed to show the advantages of the CED administered liposome carrier for use in intratumoral radionuclide therapy as compared with CED administered small molecules. At 4 h following CED administration, a significantly higher amount of ¹⁸⁶Re activity (42% of the administered dose) was retained in the tumor as compared with the naked ¹⁸⁶Re-perrhenate (only 0.2% tumor retention) and ¹⁸⁶Re-BMEDA (only 1.7% tumor retention). At 20 h, 39.7% of the ¹⁸⁶Re-liposomes were still present in the tumor whereas only 0.1% of ¹⁸⁶Re-perrhenate and 0.7% of ¹⁸⁶Re-BMEDA were retained. This difference in retention is graphically shown in Fig. 3A. Based on this prolonged tumor retention time, the radiation absorbed dose delivered to the tumor was much greater with ¹⁸⁶Re-liposomes (526 Gy) versus ¹⁸⁶Re-perrhenate (3.3 Gy) and ¹⁸⁶Re-BMEDA (13.4 Gy).

Fig. 3.

Tumor retention (A) and therapy response (B–C) of ¹⁸⁶Re-liposomes compared with free ¹⁸⁶Re-perrhenate, ¹⁸⁶Re-BMEDA or unlabeled liposome controls following intratumoral CED administration in 2-gram head and neck tumor xenografts. Tumor retention of ¹⁸⁶Re-liposomes was much greater than the free ¹⁸⁶Re-perrhenate or the small chelator molecule, ¹⁸⁶Re-BMEDA. This improved tumor retention of the ¹⁸⁶Re-liposomes as well as its improved intratumoral dispersion resulted in a significant therapeutic response that was not observed when the same amount of ¹⁸⁶Re activity was administered in the chemical form of either free ¹⁸⁶Re-perrhenate or ¹⁸⁶Re-BMEDA (***p < 0.001 for ¹⁸⁶Re-liposomes versus the ¹⁸⁶Re and liposome control groups). When the therapeutic response of individual animals to the¹⁸⁶Re-liposomes was examined, ~hal fo fthe animals appeared to have a complete response while the other half had tumors that re-grew at a much slower rate than that of control animals.

Not surprisingly, the greatly improved tumor retention and higher radiation absorbed dose delivered to the tumor by the ¹⁸⁶Re-liposomes was associated with a significantly greater therapeutic response in the group of rats infused with ¹⁸⁶Re liposomes, with virtually no observable therapeutic effect noted in the ¹⁸⁶Re perrhenate or ¹⁸⁶Re-BMEDA groups. With ¹⁸⁶Re-liposomes, there was an 87% reduction in average tumor volume with the ¹⁸⁶Re-liposomes whereas ¹⁸⁶Re-perrhenate, ¹⁸⁶Re-BMEDA or control unlabeled liposomes all had a greater than 395% increase in tumor volume as shown in Fig. 3B. Tumor responses in individual animals administered ¹⁸⁶Re liposomes are shown in Fig. 3C. Interestingly, when examined by individual treated animal, half of the tumors appeared to be completely suppressed over the 47 days of the study whereas the other half had a gradual tumor regrowth at a much slower rate than the initial growth rate of the untreated control tumors.

Several observations can be made regarding this study. First, it is important to note that the initial size of this tumor treated by CED is ~2 g which is in the size range of human tumors. Thus, CED administered ¹⁸⁶Re-liposomes are likely to have the ability to control human tumors of this size if a good CED intratumoral nanoparticle distribution can be achieved. Second, in examining the individual tumor responses as shown in Fig. 3C, it is likely that the three recurrent tumors did not receive a complete coverage of the tumor, yet the ¹⁸⁶Re-liposome treated tumors regrew at a much slower rate when compared to the initial rapid growth rate that the control untreated tumors had. A similar finding of slow recurrent tumor growth was also observed in a study by our group of ¹⁸⁶Re-liposome glioma brain tumors [142]. Although the cause of the slower tumor regrowth is not known, one possibility is that an immune response was induced. An immune response to radiotherapeutic nanoparticles administered intratumorally has been recently described with a ¹⁸⁸Re-lipid nanocapsules used for treatment of a rat glioma tumor [165]. Third, since these animals demonstrated no observable toxicity to the treatment either locally or systemically in terms of hematology orweight changes, the animals could have likely tolerated another local CED administration of the ¹⁸⁶Re-liposomes with the possibility of further prolongation of survival or cure.

Improved tumor retention was also clearly shown with another type of intratumorally administered nanoparticle, gold nanoshells. In this study by Xie et al., gold nanoshells administered intratumorally, had a 5-fold greater retention in the solid tumor at 48 h post-injection as com-pared with either an un-modified small molecule or a much larger molecular weight pegylated molecule [15]. An interesting finding of this study [47] is that the larger pegylated ⁶⁴Cu-radiolabeled molecules did not have good retention in the tumor in comparison with the gold nanoshells or in comparison with prior observations by our group of CED intratumorally administered radiolabeled liposomes [64,91,142].

These studies demonstrate the prolonged retention of nanoparticles in tumors following intratumoral CED administration. In a study in which liposome encapsulated MRI contrast agents were administered to normal brain tissue, these contrast agents cleared from the brain over a prolonged half-life of ~60 h [166]. It has been suggested that this prolonged retention is due to the uptake of liposome encapsulated agents by perivascular cells, most likely perivascular macrophages [166]. These perivascular cells are said to have relatively short half-lives as they are replenished by circulating monocytes [167], however their retention and circulation time within tumors is likely to be long enough for significant radiation to be delivered to brain tumors considering the previously observed 60 hour half -life of CED administered liposomes within brain tumor and the documented retention of CED administered liposomes within perivascular macrophages for at least 2 days [166]. However, it could also be that the nanoparticles are taken up by any of the high concentration of microglia in brain tumors which have a slow turnover [168].

In general, the high concentration of phagocytic cells in brain tumors, are likely responsible for the increased retention of CED administered nanoparticles in brain tumors, thus making these phagocytic cells very promising targets for radiotherapeutic nanoparticles. These phago-cytic cells in the brain have been termed monocyte derived cells of the brain (MDCB) [169]. It has been have shown that the higher the concentration of MDCB in glioblastoma tumors, the worse a patient’s prognosis [170]. MDCB are thought be associated with three functions within gliomas including 1) promotion of tumor invasion and growth, 2) promotion of angiogenesis vasculogenesis and 3) immune suppression. Because tumors have been shown to have a higher concentration of phagocytic MDCB than normal brain [171], tumor tissue may have a longer retention of nanoparticles than normal brain. Considering that intratumoral retention time is a very important factor affecting the total radiation absorbed dose delivered from therapeutic radionuclides, nanoparticles could act as “Trojan horses” that are phagocytized by the very MDCB that are both targeted cells and responsible for their increased retention within the tumor. In the case of radiotherapeutic nanoparticles, their 2 mm beta radiation could be delivered to surrounding stem-cells within the tumor further increasing their antitumor effectiveness.

4.2. Nanoparticles have excellent dispersion through the tumor with CED administration

Work by our group and others have shown that intratumorally administered nanoparticles have excellent dispersion through solid tumors when administered by CED [166,172]. Based on work by Bao et al. using scintigraphic imaging to track the distribution of agents directly injected into tumors, nanoparticles infused by CED can have much greater spread throughout a local tissue region while also maintaining significant retention of ~50–60% of the nanoparticles in this local infused region [64,65]. This advantageous spread of nanoparticles through a solid tumor can be seen when radiolabeled liposomes were infused directly into a solid tumor as compared to the infusion of the small molecule, BMEDA (Fig. 4) [17]. The small BMEDA molecules had decreased local spread and greatly decreased local retention which was likely due to a more rapid absorption into blood capillaries.

Fig. 4.

SPECT/CT images of rats with head and neck tumor xenografts acquired 20 h after the intratumoral administration of ^99mTc-liposomes (A, C) versus ^99mTc-BMEDA chelator (B, D). In the sagittal SPECT images superimposed on co-registered CT images (A, B), note the white color located centrally at the site of the intratumoral administration of the^99m Tc-BMEDA chelator representinga very high local concentration of activity, which indicates the poor dispersion capability of small molecules from CED administration. In contrast, the broad range distribution of radioactive liposomes shows the advantages of the use of liposome nanoparticles for improved drug distribution coverage in the solid tumor. Trans-axial SPECT images of the excised tumor intratumorally infused with ^99mTc-liposomes (Fig. 3C) or ^99mTc-BMEDA (Fig. 3D) also show the improved distribution of ^99mTc-liposome nanoparticles.

In the case of brain cancer, most of the nanoparticle distribution studies have focused on the dispersion in normal brain tissue. Magnetic resonance imaging (MRI) of liposomes has shown that the volume of liposomes infused by CED into the normal brain tissue is highly correlated with the total brain tissue distribution volume. In these studies, the brain distribution volumes are approximately 3-fold greater than the volume of liposomes infused [173].

Interestingly, positively charged cationic liposomes, that were shown to have a 50 fold higher binding constant for normal brain tissue had approximately 3.5 times shorter dispersion distance in tissue as compared with neutral and negatively charged liposomes [166]. Based on a review of previous studies, Allard et al. proposed optimal nanoparticle properties including a particle size of <100 nm, neutral or negative surface charge and a steric coating to reduce binding of the nanoparticles to brain cells [152].

5. History of intratumorally administered radiotherapeutic nanoparticles

Prior to the development of nanoparticle-based therapeutic radionuclide agents, a moderate amount of research was performed with micron-sized particles carrying β-emitting radionuclides administered intratumorally [58,125]. These early studies with β-emitting particles are instructive regarding the potential for nanoparticle radionuclide therapy and they have been previously reviewed by McCready and Cornes [125]. Chromic phosphate particles (600–1300 nm in diameter) radiolabeled with the β-emitting radionuclide, phosphorus-32 (³²P), was one of the first particles investigated for direct intratumoral radionuclide therapy. ³²P has a relatively long half-life of 14.29 days and a maximum β-emission energy of 1.7 MeV giving it an average tissue path length of ~3 mm. The large and variable size range of ³²P-chromic phosphate is due to the tendency of the colloidal particle to aggregate and form particles in the micron size range in a relatively uncontrolled fashion. Particles of this size range are not likely to have a significant amount of dispersion distance through the tumor when injected intratumorally; however, they are likely to have good intratumoral retention. In fact, the large ³²P-chromic phosphate particles were chosen by the investigators because in preliminary studies, the ³²P-chromic phosphate particles showed “virtually no diffusion from infiltrated regions to non-infiltrated regions”, emphasizing tumor retention over penetration within the tumor.

In spite of this minimal dispersion distance, these initial studies with ³²P-chromic phosphate in humans demonstrated very promising results for image-guided cancer therapy in patients. In a Phase II study of 17 patients by Firusian et al. [60], patients with chemotherapy-resistant tumors were intratumorally administered ³²P-chromic phosphate by ultrasound image-guidance. An encouraging response rate of 71% was reported with complete remission in 41% of the patients [60]. A wide variety of tumors were treated in this study including lymph node metastasis from adenocarcinoma of the breast, squamous cell carcinoma of the lungs, metastasis of the liver from adenocarcinoma of the stomach and colon, and primary hepatocellular carcinoma. Minimal side effects of the treatment were noted. Among the 17 cases treated, thrombocytopenia was noted in 1 patient that was treated with 3 doses of ³²P-chromic phosphate. In 2 other patients with superficial lymph node metastasis, transient erythema and a burning sensation without late effects was noted [60]. Interestingly, many of the treated tumors in this study developed a central cystic necrotic region which the authors attributed to the “unavoidable inhomogeneous accumulation of activity.”

In a follow-up article, this group reported a study of 14 patients with refractory head and neck cancer treated with intratumoral ³²P-chromic phosphate particles [61]. Eight of the 14 patients had a partial response to treatment associated with remarkable regression. Follow up pathologic investigation of treated tumors demonstrated a central cystic lesion surrounded by a centrifugal necrotic region and a marginal fibrotic structure. Three of the treated patients developed grade I/II thrombocytopenia. It is likely that this thrombocytopenia is related to the avidity of ³²P for bone and that the ³²P-chromic phosphate particles were degraded to free ³²P and then depositing in bone adjacent to active bone marrow platelet formation. Overall, this initial work with intratumorally administered β-emitting radionuclide carrying particles was considered successful in patients with chemotherapy resistant tumors and this work brought attention to the potential of intratumoral radionuclide particle therapy.

Table 2 reviews previous work with intratumorally administered nanoparticle carriers of radionuclides in which studies have been performed.

Table 2.

Previous research with intratumorally administered radiotherapeutic nanoparticle.

Agent	Size (nm)	Tumor treated	Species (number)	Radio activiy mCi (MBq)	Results	Toxicity	Reference
P-32 chromic phosphate colloid	600–1300	Pancreatic	Human (N = 28)	4–60 (148–2220)/tumor	Median survival − 12 months without metastasis and 6.9 months with metastasis	Grade IV thrombocytopenia (2 patients) Grade III neutropenia (1 patient) No significant toxicity when dose limited to 2 infusions of 30 mCi each	[58]
P-32 chromic phosphate colloid	600–1300	Variety of solid tumors	Human (N = 17)	2–15 (74–555)/ tumor	Therapy response − 71% Complete response − 41%	Minimal	[60]
P-32 chromic phosphate colloid	600–1300	Head and neck	Human (N = 14)	2–12 (74–444)/ tumor	Partial response − 51%	Grade I/II thrombocytopenia (3 patients)	[61]
Au-198 gold nanoparticle (epigallocatechin-gallate targeted)	120–130	Prostate xenograft	Mice (N = 7)	0.136 (5)/tumor	80% decrease in tumor volume at Day 28	No clinical signs	[50]
Au-198 gold nanoparticle (Gum Arabic glycoprotein coated)	85	Prostate xenograft	Mice (N = 7)	0.408 (15)/tumor	82% decrease in tumor volume at day 21	Transient weight loss No systemic toxicity	[49]
Au-198 gold dendrimer nanodevice	10–29	Mouse melanoma	Mice (N = 14)	0.354 (13)/tumor (N = 7) 0.0744 (2.75)/tumor (N = 7)	45% decrease in tumor volume at day 8	No clinical toxicity	[48]
Lu-177 metallofullerene	20	Glioblastoma intracranial xenograft	Nude mice (N = 10)	0.03 (1.11)/tumor	At day 52 90% of treated mice survived compared with no control mice	Not determined	[54]
Re-188 nanocapsules	55.5 ±11.5	Rat 9L gliosarcoma	Rat (N = 19)	0.03–0.12 (1.1–4.4)/tumor	For 2.8 MBq group, 33% of rats survived to day 100, an increase in median survival of 80% compared with control groups.	Side effects of lethargy and 25% weight loss observed in 4.4 MBq group. Slight body weight loss observed in 2.8 and 3.7 MBq groups until day 12.	[69]
Re-186 liposomes	135.1 ± 18	Head & neck xenograft	Nude rat (N = 6)	5 (185)/cm3 tumor	87.7% decrease in tumor volume on day 42	Mild body weight loss until day 5 No systemic toxicity observed	[64]
Re-186 liposomes	109 ± 11	Head & neck xenograft reminant	Nude rat (N = 8)	4.75 (176)/rat	Remnant tumor volume increased 25.6% compared with 288% increase for unlabeled liposome control.	Mild body weight loss until day 5 No systemic toxicity	[146]
Re-186 liposomes	108 ± 26	Glioblastoma intracranial xenograft	Nude rat (U87 N = 18; U251 N = 5)	0.025–0.125 (0.92–4.6)/tumor	Median survival for U87 treated group was 126 days compared with 49 days for unlabeled liposome control group. Median survival was not reached at 120 days for U251 treated group.	No clinical or microscopic toxicity	[142]

5.1. Liposomes

5.1.1. Head and neck cancer

The potential use of nanoparticles for delivery of therapeutic beta-emitting radionuclides intratumorally was first described by Bao et al. in 2006 (Fig. 3) [65]. This article used the diagnostic imaging agent, technetium-99m (^99mTc), to study the distribution of intratumorally administered liposomal nanoparticles by SPECT/CT imaging in a rat tumor model. This article describes the promising intratumoral distribution and retention of the ^99mTc-labeled liposomes within solid tumors and pointed out the promise of liposomes as carriers for the therapeutic β-emitting radionuclides of rhenium including ¹⁸⁶Re and¹⁸⁸Re. In a previous article published in 2003, Bao et al. had already shown that the same chemistry for labeling liposomes with ^99mTc was also effective for labeling liposomes with ¹⁸⁶Re [66]. Intratumorally administered liposomes carrying radionuclides had significantly improved intratumoral dispersion and retention as compared with the lipophilic small molecule, BMEDA. Based on this work, liposomes were shown to be ideal candidates for CED administration of therapeutic radionuclides in addition to other drugs. Liposomes and lipid nanoparticles are by nature bio-tolerable, due to naturally occurring metabolic pathways for lipids.

In a follow-up study in 2008, the author’s group showed therapeutic efficacy of 100 nm ¹⁸⁶Re-liposomes in an intraoperative model of head and neck cancer [146]. This model was designed to treat the situation of residual tumor following cancer surgery. Rats with an initial tumor size of >3 g had a majority of their tumor removed leaving an ~1 gram tumor remnant which was intratumorally infused with ¹⁸⁶Re-liposomes as compared with a sham surgical control administered an unlabeled liposome formulation. At the end of the study at 35 days, the ¹⁸⁶Re-liposome-treated tumors were ~25% of their post-surgical baseline volume, while the control tumors were 290% of their baseline volume. A total of 8 animals were treated with ¹⁸⁶Re-liposomes; 4 treated with neutral liposomes and 4 with cationic liposomes. Both types of liposomes appeared equally effective and at the end of the study, all animals had smaller tumors than they had at post-surgical baseline. There was no significant difference in weight gain or white blood cell counts in the treated animals as compared with the control animals. This study was the first to show therapeutic efficacy of intratumorally administered radiotherapeutic liposomes for solid cancers.

Intratumoral CED therapy of a solid head and neck tumor model with ¹⁸⁶Re-liposomes was previously described in Section 4.1 in regard to intratumoral retention.

Excellent results were found using ¹⁸⁶Re-liposomes administered by CED to head and neck tumor xenografts of 2 g in size at the time of initial therapy (see Fig. 3).

5.1.2. Glioblastoma

In 2012, our group showed promising results in the use of CED administered ¹⁸⁶Re-liposomes for the treatment of glioblastoma in U87 rat glioma model and in the more aggressive U251 rat glioma model [142,174]. Glioblastoma is an ideal target for local nanoparticle radionuclide therapy due to its poor prognosis with current therapies and local uncontrolled growth as the cause of death with 90% of tumor recurrences occurring within 2 cm of the resection. In the initial portion of the study, the ability to perform non-invasive diagnostic imaging was very useful for determination of the volume and rates to be used for the intratumoral CED infusion. Follow determination of ideal CED parameters, a wide variety of ¹⁸⁶Re-liposome radiation absorbed doses were examined and a significant therapeutic benefit was noted in U87 tumors with doses of greater than 100 Gywith >50% of the animals surviving until the end of the study at 120 days. No significant toxicity was noted in doses up to 1500 Gy. Serial MRI images clearly demonstrated the therapeutic effect of the ¹⁸⁶Re-liposomes particularly in comparison with control images as shown in Fig. 5A.

Fig. 5.

MRI-based therapy response monitoring (A) and survival (B) of rats with intracranial glioblastoma xenografts after treatment with ¹⁸⁶Re-liposomes. MRI images of intracranial tumors reveal comparable pretreatment U87 brain tumor size in control animals and in animals treated with intratumoral CED administration of ¹⁸⁶Re-liposome nanoparticles. Tumors in control animals had progressive growth by day 14 post-tx , whereas ¹⁸⁶Re-liposome-treated tumors gradually decreased over time and disappeared at last time point studied. Survival was greatly extended for the rats treated with ¹⁸⁶Re-liposomes. Four of the five animals treated by CED infusion of ¹⁸⁶Re-liposomes showed no evidence of U251 tumor re-growth at the end of the study, approximately 100 days post-treatment. All control animals died by 17 days after the time of the start of treatment in the experimental animals.

In a follow up of the more aggressive U251 tumor, rats treated with ¹⁸⁶Re-liposomes, carrying a mean dose of 1700 Gy, were compared with control rats (Fig. 5B). No control rats survived past 35 days post tumor implantation while 83% of the treated rats were still alive at 130 days post implantation. Histology did not reveal U251 tumor cells in any of the surviving rats.

These pre-clinical results in this aggressive glioma model show the promising potential of CED administered ¹⁸⁶Re-liposomes for glioblastoma therapy. Based on extensive pre-clinical work with ¹⁸⁶Re-liposomes demonstrating their efficacy for solid tumor therapy, ¹⁸⁶Re-liposomes appear to be promising therapeutic agents for image-guided CED treatment of glioblastoma and the local control of other types of solid tumors.

5.2. Solid lipid nanocapsules

In 2008, Allard et al. described the use of solid lipid nanocapsules as carriers for the therapeutic radionuclide, ¹⁸⁸Re, for the treatment of malignant gliomas [69]. ¹⁸⁸Re has a β-emission with twice the energy of ¹⁸⁶Re and an average tissue path length of 4 mm. Rats with 9L brain tumors were treated with ¹⁸⁸Re-nanocapsules at 6 days after implantation. The survival time was increased by 80% in the treated rats as compared with the control rats with 33% of the rats being long-term survivors. This study was the first to show the in vivo efficacy of ¹⁸⁸Re internal radiotherapy for the treatment of brain malignancy.

5.3. Radioactive gold nanoparticles

In 2008, Khan et al. described the use of an intratumorally administered radioactive gold-198 (¹⁹⁸Au) dendrimer composite nanodevice in the treatment of a mouse melanoma model [48] ¹⁹⁸Au has a half-life of 2.7 days and an average β-radiation tissue path length of 2 mm, similar to ¹⁸⁶Re. Eight days after treatment, there was a statistically significant 45% reduction in tumor volume when compared with untreated groups and those injected with the “cold” nanodevice. No toxicity was observed in the treated mice.

In 2010, radioactive gold-198 (¹⁹⁸Au) nanoparticles were investigated in a model of prostate cancer [49]. These gold nanoparticles had a hy-drodynamic diameter of 85 nm.

On day 8 after implantation in the flank of mice, tumors were treated with ¹⁹⁸Au gold nanoparticles and tumor growth was compared to control animals receiving injections of phosphate buffered saline. Three weeks after treatment, the tumors of the treated animals were 82% smaller in comparison with the tumor volume of the control animals. At the end of the 3-week period, 19% of the initially infused ¹⁹⁸Au still remained in the tumor. On pathological examination, the tumor consisted mostly of necrotic tissue, indicating extensive tumor cell killing. This group also functionalized the ¹⁹⁸Au-nanoparticles by adding epigallocatechin-gallate (EGCg) to the surface of the¹⁹⁸Au-nanoparticles to specifically target the laminin receptor which is over-expressed on prostate cancer cells. In a mouse prostate cancer model, there was an 80% reduction in tumor volume at 28 days after treatment with the intratumorally administered EGCg-¹⁹⁸Au-nanoparticles [50]. A recent review of the investigations of intratumorally administered functionalized radioactive gold nanoparticles in tumor therapy has been published [175].

5.4. Metallofullerene nanoparticles radiolabeled with lutetium-177 (¹⁷⁷Lu)

In 2011, a multimodal β-emitting metallofullerene-based nanoplatform was developed for brain tumor brachytherapy. This nanoplatform carried the β-emitting radionuclide of ¹⁷⁷Lu [54] while also carrying the MRI contrast agent, gadolinium. ¹⁷⁷Lu has an average β-radiation tissue penetration distance of 0.7 mm and a radionuclide half-life of 6.7 days. When administered by CED to mice with intracranial U87 brain tumors implanted 8 days previously, the treated mice had survival times of >2.5 times the control untreated mice. At the end of the study at 60 days, small but viable tumor remained present in the brain of the treated mice; however, the small size suggests that the growth rate of the tumor was very significantly slowed. Immediately after CED infusion, 60% of the infused dose remained in the region of the brain tumor. Follow-up MRI images demonstrated an extended retention of the metallofullerenes within the tumor region with 25% of the administered metallofullerenes retained at 52 days [54]. In contrast, free chelated gadolinium injected intratumorally cleared from brain rapidly with <1% of the infused dose retained in the brain at 24 h. The size of these metallofullerene particles in vivo was not easily determined as in vitro studies of their particle size ranged from 1 to 20 nm depending on the pH of the medium with larger clustering of metallofullerenes observed at lower pH. One of the limitations of this study noted by the authors was that there was no comparison of the therapeutic efficacy of the ¹⁷⁷Lu-1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) chelator alone, however based on the studies by our group, it is likely that ¹⁷⁷Lu-DOTA administered alone would not be effective for brain tumor therapy due to its rapid clearance from the tumor [47,64]. A second limitation of the study is that the initial tumors were very small at the time of treatment. At 8 days post implantation, tumors were not well visualized on the baseline MRI images and represent less of a therapeutic challenge in comparison with larger brain tumors.

6. Future directions for image-guided nanoparticle therapy

6.1. Development of methods to improve CED coverage of and retention within a solid tumor

Future research will be needed to investigate convenient and effective methods to achieve complete coverage of a solid tumor. Using current CED technology, complete coverage of the tumor with CED infused nanoparticles will likely require that multiple infusions be administered at spaced intervals within the same tumor. Based on current work by the authors using standard CED systems, nanoparticle tissue coverage from CED is estimated at ~1–2 g of tumor tissue. Current research is being conducted to increase CED tissue coverage using a variety of techniques [128,140,143,176–179].

6.1.1. Improvement in catheter design

In the last decade, there has been significant improvement in the design of CED catheters to allow greater CED coverage of an infused tumor. These improvements included the use of catheters with tips of less than 1 mm in diameter to decrease tissue trauma [180], and the use of stepped and recessed step catheters to facilitate high flow infusions with great volume of distributions and eliminate backflow along the catheter track [140]. However, despite these advancements, there is still much room for further improvements in catheter design such as the development of multiple pore and multipronged catheters that will cover larger tumor volumes from a single infusion.

6.1.2. Use of non-invasive image guidance to improve solid tumor coverage

6.1.2.1. Advanced imaging for monitoring intratumoral nanoparticle distribution and tumor response

To date, only a few studies have examined in a quantitative fashion the intratumoral retention and intratumoral distribution of nanoparticles [64,65,142,146]. Non-invasive imaging is ideally suited to monitor tumor retention and distribution of CED administered agents and to correlate tumor response to the delivery and retention of nanoparticles in individual patients. With radiotherapeutic nanoparticles which emit gamma photons as well as therapeutic beta particles, it will be possible to correlate the distribution and retention of intratumorally administered agents with the therapeutic response of the tumor in individual patients. In the future, it may be possible to determine the distribution and therapeutic dose delivered to a treated tumor in real time using advanced imaging techniques.

In addition, physiologic imaging of the tumor response to CED administered radionuclide therapy can be done using PET imaging. Rapid assessment of tumor treatment efficacy is possible with a variety of PET radiopharmaceuticals. For example, markers of cell division such as fluorine-18 (¹⁸F)-fluorothymidine could provide feedback, feasibly within a1-week time period following the intratumoral CED nanoparticle radiotherapy. An advantage of ¹⁸F-fluorothymidine is that it specifically demonstrates regions in the tumor of rapid cell division; and unlike, ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) is not increased typically in sites of inflammation. With the advent of clinical PET/MRI high resolution imaging scanners, focal regions of residual tumor could be retreated with specifically targeted image-guidance and without fear of local toxicity, which is a major advantage of β-emitting radionuclides due to their limited path length of the β-particle radiation.

Image-guidance is likely to play a very important role in the placement and real-time changes in the location of the CED catheter with a tumor undergoing treatment with nanoparticles that can be imaged. The best location of placement for the CED treatment of solid tumors is not known due to very little research that addresses this problem. Because the intratumoral pressure is highest in the center of the tumor which tend to move nanoparticles out towards the tumor rim, the central regions of the tumor would appear to be an important targeted region for CED infusion. Most of the current research with CED has been performed in small animal models of normal brain or with tumors, and which is not likely to translate well to human studies. Restriction of this prior work to small animals is likely due to the minimal availability of large animal brain tumor models [141,148]. It is the author’s belief that the best approach will be to use real-time imaging, likely MRI, to ensure that nanoparticles have covered as much of the tumor as possible, including the core and rim of the tumor, until more knowledge is gained from future clinical real-time intratumoral distribution studies. Because current distribution patterns are not entirely predictable, most researchers have emphasized the importance of real-time image guidance that adjusts the placement of the CED infusion during the treatment so as to paint as much of the tumor as possible [129,141, 143,147].

6.1.2.2. Prediction of nanoparticle distribution based on tumor microstruc-ture as determined from prior high resolution anatomic imaging

High resolution MRI scanners with very high strength magnetic fields have been recently described for use in brain imaging, providing micro-anatomic details and an in-plane resolution of 0.23 mm in human brain [181]. It may be possible to use this high resolution, anatomically detailed MRI imaging of tumor microanatomy to plan and predict distribution of CED administered nanoparticles within tumors. Extensive pre-therapy planning could be performed to determine the precise locations of catheter placements and infusion volumes and rates using computer algorithms. Actual distributions could be monitored with real-time high resolution MRI [143] with the possibility of on-the-fly adjustment of CED infusions to ensure total tumor volume coverage.

6.1.2.3. Development of treatment planning systems for CED catheter planning and real-time imaging to monitor CED tumor coverage

To achieve full tumor coverage, planning systems using computer based algorithms could be developed to pre-select the site of CED infusions based on known CED volumes of distribution and potentially using microstruc-tural information of the tumor obtained from high resolution MRI imaging prior to treatment. Feasibility of this real-time approach has already been shown in primate brain using liposomes carrying MRI contrast agents [143]. This liposomal MRI contrast made it possible to monitor the intratumoral distribution of nanoparticles administered by CED using MRI. Nanoparticles carrying both MRI contrast and β-emitting radiotherapeutic isotopes could be administered under intraoperative MRI-guidance in real-time to ensure the proper distribution within the tumor. This real time application of MRI image guidance for CED administration could ensure that the entire tumor is covered with radiotherapeutic nanoparticles. This real-time image-guidance is similar to the current use of advanced MRI and PET/CT for “dose painting” by external beam radiation therapy [182].

6.1.2.4. Potential for image-guided robotically administered CED

Radiotherapeutic nanoparticles have potential to be delivered into solid tumors by robots using image-guidance to achieve complete tumor coverage. Robotic placement of catheters using image guidance has recently been described as a method for obtaining brain stem biopsies with promising results based on image-guidance and frameless stereotactic localization of the biopsy needle by the robot [183]. In one study, 15 patients (five children and ten adults) underwent biopsies of brain stem lesions using a frameless image-guided robot. The biopsy procedure required four stages: image acquisition, preoperative planning, patient-to-image registration, and operative procedure. There was no operative mortality. The investigators concluded that the use of frameless stereotactic techniques for brain stem biopsies could increase the number of biopsies and therefore improve the diagnostic yield and accuracy of the technique.

Feasibility of using robots for delivering a CED therapy has also recently been demonstrated. In this study, robotic delivery of carboplatin into a diffuse pontine glioma was performed [179]. This procedure used stereotactic planning software to develop the catheter placement approach. In this procedure, a transfrontal trajectory for catheter implantation was planned for positioning of a recessed-step catheter. The catheter was implanted using a specially designed robot for brain catheter placement. Carboplatin was infused into the tumor while real-time T2-weighted MRI was performed, facilitating estimation of the total volume of infusate distribution [179].

6.2. Further investigations into the ideal characteristics of nanoparticles for CED with enhanced volume of distribution and tumor retention

More in-depth knowledge is required regarding the effect of nanoparticle size on the volume of distribution achieved during CED administration. How the size of the nanoparticle relates to the volume of distribution, and perhaps, more importantly, the retention of the nanoparticle within the tumor are areas for further research. Also, the effect of surface charge and the addition of specific surface receptors on the CED volume of distribution are thought to be important parameters to consider. In a study by Szoka’s group, liposomes with cationic surface charge had less dispersion distance than liposomes with neutral or negative surface charges [73]. Initial studies examining these characteristics have been performed, but further more detailed studies are needed. Non-invasive imaging in clinical studies with MRI and nuclear imaging (PET and SPECT) can be used to examine different nanoparticle characteristics on volume of distribution and retention in targeted tumor tissue.

6.2.1. Investigations of the most ideal size of nanoparticles for CED administration

Only a few studies have been performed specifically looking at the most ideal size of nanoparticles for CED drug carriers. Theoretically, the size of the nanoparticle would be smaller than the pore size of the extracellular matrix of the tissue being treated. The majority of the research investigating extracellular matrix pore size has been for brain drug delivery. The pore size of the extracellular matrix of normal brain extracellular space has been reported to be 38–64 nm in size [184] and the pore size of the extracellular space of brain tumor was found to be larger at 70–100 nm in size [185]. Zhou et al. have suggested that nanoparticles for the treatment of brain tumors by CED should be 60–80 nm in diameter to allow access to the brain tumor while sparing the normal brain [185]. Zhou etal. developed a 74 ± 18 nm PLGA nanoparticle and compared its CED volume of distribution to 150 ± 30 nm diameter PLGA nanoparticle in normal rat brain [186]. The mean volume of distribution of the small 74 nm nanoparticles was ~7 times greater than the 150 nm nanoparticles. These 74 nm nanoparticles were also studied in a normal pig brain model and the volume of distribution (Vd)/volume of infusion (Vi) ratio was 3.5 with a total volume of distribution of 1.18 cm³ which was considered to be clinically relevant due to the fact that most glioblastomas recur within 2 cm of their original location.

The type of nanoparticle may also play a role in their distance of penetration by CED. For example, liposomes which have some deformability could potentially squeeze through pores without losing their integrity. In a study investigating different types of CED administered liposomes, liposome nanoparticles of up to 200 nm in size and of neutral or negative charge were shown to have a similar radius of penetration when injected into normal rat brain [166]. No significant differences in the radius of penetration were noted for neutral liposomes of 85 nm in diameter or negatively charged liposomes of 83 nm in diameter. Neutral liposomes of 197 nm had a slightly smaller radius of penetration than the 85 nm liposomes. Nanoparticles larger than 200 nm began to be retained near the point of infusion. In this study, 10 kDa dextran molecules penetrated significantly further in tissue as compared to 83 nm liposome nanoparticles when administered by CED, however their time of retention in the tumor was not assessed [166].

Although studies are limited, it appears that tissues outside of the brain may have significantly large extracellular space pores. Although not the same as extracellular space pores, the vascular pore cutoff sizes in tumors grown subcutaneously (200 −1,200 nm) was much larger compared to 100–380 nm for intracranially grown tumors [187]. The possibility of substantially larger pore cutoff sizes in the extracellular matrix of tumors outside the brain could make image-guided delivery to these tumors more rapid and easier to accomplish. In a rat head and neck tumor, the CED infusion rate was 0.5 ml/min which is 100– 200 times more rapid than in the typical CED infusion protocols in brain CED applications. In this head and neck model, non-invasive imaging demonstrated fairly homogeneous distribution throughout the tumor. Certainly, much more investigation of nanoparticle CED applications in tumors other than brain are needed.

6.2.2. Multifunctional nanoparticles for tracking CED distribution within tumors

An article by Li et al. and authors of this paper have shown that the same liposome can serve as a carrier not only for the therapeutic radionuclides of rhenium, ¹⁸⁶Re and ¹⁸⁸Re, but simultaneously also for a therapeutic drug, as well as 4 different types of imaging contrast agents including an MRI contrast agent, PET and SPECT nuclear diagnostic imaging agents and optical contrast agents as shown in Fig. 6 [160]. The ability to perform multiple types of imaging on the same nanoparticles will allow studies investigating the distribution and retention of nanoparticles initially in vivo using non-invasive imaging and later at the histological level using optical imaging.

Fig. 6.

Multimodal MRI (A), PET (B), near infrared optical (C) images of head and neck tumor acquired following intratumoral CED administration and schematic diagram of multifunctional liposomes demonstrate the versatility of nanoparticles for multifunctional imaging and therapy applications. These multiple images were obtained from the same liposome nanoparticle formulation. Intratumoral distribution of the multifunctional liposome nanoparticles was non-invasively acquired with MR imaging through gadolinium-DOTA (A), PET imaging with ⁶⁴Cu-labeling on DOTA group (B), and optical imaging with IRDye on headgroup (C). Multifunctional liposomes can be very useful in pre-clinical research and clinical applications by providing different means of monitoring the distribution and retention of the nanoparticles. The strength of MR imaging is its ability to provide high resolution distribution non-invasively. The strength of PET imaging is its ability to provide quantitative data over time on the distribution and retention of nanoparticles within the tumor. The strength of near infrared optical imaging is in its use for non-invasive imaging in small animals, its potential use in real-time surgical applications and its use for distribution studies at the optical microscopic level using histological slides.

6.3. Therapeutic ultrasound to enhance the convection of intratumoral nanoparticles

One promising approach under investigation is the use of therapeutic ultrasound-assisted CED [178]. With this approach, the vibration of the ultrasound probe during convection enhanced delivery results in an enhanced rate of intratumoral administration and a larger volume of distribution of infused agents. Studies in the rat brain using a novel, low profile ultrasound transducer positioned in the brain along with the CED catheter demonstrated that the volume of the CED infusion could greatly increase the volume of distribution of Evans Blue dye by ~3-fold in normal brain. From histological examination of the brain, no changes were noted that could be attributed to the perturbation from ultrasound application. Certainly, further investigations of this promising CED delivery technique are needed. The challenges for this ultrasound technology and CED cancer therapy in general were concisely addressed in a commentary by J.H. Sampson Jr. [148].

6.4. Investigations into different β-emitting radionuclides for intratumoral image-guided therapy

Further investigations can be performed with different radionuclides for potential use in radionuclide nanoparticle therapy. For different applications, different tumor types and different tumor locations in the body, there may be specific advantages in using radionuclides with different half-lives and different beta energies. For example, the shorter half-life and longer tissue path length of ¹⁸⁸Re may have specific advantages for tumor treatment where the longer tissue path length will prove advantageous. Liposomes have already been shown to be stably labeled with ¹⁸⁸Re by Chen and Chang et al. [73,188,189] who have reported promising results of ¹⁸⁸Re-liposomes for intravenously administered anti-tumor applications. There could also be important cost advantages to specific radionuclides. For instance, the majority of radionuclides for therapy are made from a reactor and used in a fairly short time frame of less than a week. The rapid use is associated with significant production cost and expedited shipping. Other therapeutic radionuclides are produced using generator technology. ¹⁸⁸Re (half-life of 17 h, tissue path length of 4 mm) can be produced from a tungsten-188 (¹⁸⁸W)/¹⁸⁸Re generator similar to the current production of ^99mTc for diagnostic imaging [190]. The availability of generator-produced therapeutic β-emitting radionuclides could significantly lower the cost of radionuclide nanoparticle therapy. The ¹⁸⁸W/¹⁸⁸Re generator could be used for up to 6 months with doses capable of treating many patients being eluded from this generator every 2 days. This radionuclide also has a photon with an ideal SPECT imaging photon energy of 155 keV. The low cost of this small, highly portable ¹⁸⁸W/ ¹⁸⁸Re-generator for use in tumor therapy with liposome nanoparticles on a per-patient basis could make radionuclide nanoparticle cancer therapy feasible in locations without access to advanced technology. Although this therapy is ideally delivered with MRI and PET image-guidance, the recent advancements in less expensive ultrasound imaging could also be used for radiotherapeutic nanoparticle therapy and palliation of ultrasound accesible tumors. Diagnostic imaging ultrasound for image-guidance delivery in cancer chemotherapy has been recently described [191].

6.5. Nanoparticles as ideal carriers to target intratumoral macrophages/microglia

Many tumors have been shown to have high concentrations of macrophages and in many tumor types, it has been shown that higher macrophage concentration within tumors is associated with a poorer prognosis [170,192–196]. In recent years, macrophages located within the tumor have become recognized as important cells supporting tumor growth [169,197–199], and this information has led to macrophages becoming important targets for the treatment of solid tumors including gliomas [198]. Because nanoparticles are prone to opsonization by blood and other body fluids [200], they serve as natural carriers that target macrophages due to the inherent phagocytic response of macrophages to particles [200]. It is well known that the majority of nanoparticles are taken up by macrophages in the liver, spleen and bone marrow following intravenous nanoparticle administration. Although not well studied, nanoparticle uptake by intratumoral macrophages is also highly likely to occur following intratumoral nanoparticle administration.

Very few investigations have been performed to determine the ultimate cellular location of nanoparticles following intratumoral CED administration. Only one recent study related to cellular location could be found by the authors in which 223 nm and 193 nm PLGA nanocapsules were administered intratumorally [201]. Histological examination at 48 h post-intratumoral infusion demonstrated that the PLGA nanoparticles were located exclusively in intratumoral macrophages.

6.5.1. Macrophage/microglia in glioblastoma tumor

Glioblastoma tumors are heavily infiltrated by cells of microglia and macrophages located in the tumor stroma, termed monocyte derived cells of the brain (MDCB). These MDCB can comprise up to 30% of the total glioblastoma tumor mass [202]. Increasing evidence has also shown that MDCB assist in glioma cell growth, motility and invasion [203]. MDCB have been shown to be closely associated with a subpopu-lation of glioma cells known as glioma stem-like cells (GSLCs) [22]. Recent studies show that these GSLCs are much more efficient at forming tumors in vivo and are likely to be substantially more invasive than the bulk of tumor cells [204]. GSLCs also appear to be the most resistant to radiation therapy [205].

A strategy to target microglia/macrophage stromal cells has been suggested to improve glioblastoma therapy [203]. An advantage of targeting the microglia/macrophages within glioblastoma is that they are genetically normal cells that have undergone a phenotypic transformation that under normal circumstances would be physiologic which would make them less likely to develop resistance to a therapy that targets them [203].

Nanoparticles carrying beta-emitting radionuclides have particular promise for effectively delivering therapy to both MDCB and associated GSLCs. Phagocytic MDCB in the glioblastoma tumor stroma are highly likely to phagocytose intratumorally delivered beta-emitting nanoparticles and radiation therapy would not only be delivered to these MDCB but the beta-particle radiation traveling through 2–4 mm of tumor tissue will also treat GSLCs. Because the local dose delivered can be much higher than with standard external beam radiation therapy, radiation resistance of GSLC would be overcome by the very high local dose delivered. In previous studies with ¹⁸⁶Re-liposome nanoparticle administered intratumorally, the dose of radiation delivered to the tumor was >1500 Gy [142]. A high radiation dose delivered to MDCB and GSLCs may provide one explanation for the surprising effective results obtained in the rat glioma tumor models with >80% of treated animals having long term survival with no residual evidence of tumor at the termination of the study [142].

6.5.2. Macrophages/microglia as intratumoral transporters of nanoparticles

The high macrophage concentration in tumors, particularly along the growing margins of the tumor may be an important component of the natural retention of nanoparticles in tumors following their CED administration, thus providing a mechanism administered liposome nanoparticles within tumor as compared to normal brain (unpublished observation by our group) [206]. It is also possible that macrophages within a tumor could carry intratumorally administered nanoparticles to the most active growing regions of the solid tumor, including the invasive margins of the tumor. This possibility is graphically shown in Fig. 7.

Fig. 7.

Illustration of the proposed mechanism of intratumoral macrophage involvement in β-radiation treatment with nanoparticles These 3 images illustrate a possible mechanism by which nanoparticles could be carried within the tumor by intratumoral macrophages to reach the rapidly growing invasive margins of the tumor. In A, the nanoparticles administered intratumorally by CED results in dispersion through the tumor. In B, dispersed nanoparticles are phagocytized by intratumoral macrophages in the dispersed locations throughout the tumor. Panel C shows migration of mobile macrophages to regions at the invasive margins of the tumor while delivering therapy to a 3D field via β-radiation treatment surrounding the macrophage.

Intratumoral macrophages could take up the CED administered nanoparticles and serve as a carrier to transport the nanoparticles to actively growing sites in the tumor.

This possibility merits further research to determine if local intratumoral macrophage transport is occurring. Macrophages have already been proposed to be carriers of therapeutic agents by preloading macrophages ex vivo with therapeutic agents and administering them intravenously [207]. This approach assumes mobility of the macrophages within the tumor. The loading of intratumoral macrophages with nanoparticles may be a simpler approach to achieve the same goal.

Certainly, more detailed investigations of the location of intra-tumorally administered nanoparticles over time post CED infusion will be important to develop a better understanding of intratumoral nanoparticle therapies, particularly those using β-emitting radionuclides. A significant advantage of this macrophage or tumor phagocytosis in regard to β-emitting radiotherapeutic nanoparticles, are that the physical location of the β-emitting radionuclides does not require release from the interior of the macrophage for effective treatment of a 40–100 cell diameter sphere of tumor cells surrounding the phagocytic cell and radiation damage to these phagocytic cells is not instantaneous, permitting time, likely days, for the macrophage to move within the tumor after phagocytosis of the β-emitting radionuclide nanoparticle. Evidence for the prolonged survival of tumor cells and likely associated macrophages is provided by a recent study using ¹⁸⁶Re-liposomes to treat an intracranial glioma xenograft. In this study, optical methods were used to non-invasively assess the viability of luciferase transfected-glioma cells following ¹⁸⁶Re-liposome intratumoral treatment (Fig. 8A). After a mean absorbed dose of 1700 Gy of ¹⁸⁶Re-liposomes, tumor cell death gradually occurred over a 12 day period (Fig. 8B). Considering the multi-day half-life of most β-emitting radionuclides, the slow death of tumor cells and tumor associated macrophages, beta-emission therapy could be transported by intratumoral phagocytic cells to the most rapidly growing tumor regions for an extended period prior to the death of the macrophage from β-radiation therapy.

Fig. 8.

Bioluminescence optical imaging can be used to detect viable luciferase gene transfected U251 tumor cells non-invasively (A) and quantify tumor cell luciferase activity in the brain of rats (B) following intratumoral CED administration of ¹⁸⁶Re-liposomes. Note that tumor cell death in response to β-emission radionuclide therapy is a relatively slow process which provides time for extensive movement of intratumoral macrophages that have ingested the ¹⁸⁶Re-liposome nanoparticles to regions of active growth in the tumor. This slow tumor death also suggests the possibility that intratumoral treatment with ¹⁸⁶Re-liposome nanoparticles could induce an immune response to the tumor.

6.6. Potential for combination of chemotherapy with radionuclide nanoparticle therapy, i.e. chemo-radionuclide therapy

As described previously regarding image-guided RF ablation, studies with intravenously administered liposome nanoparticles containing both ¹⁸⁶Re and doxorubicin demonstrated an improved tumor response in a large solid tumor model in rats as compared to liposomes containing only doxorubicin [86,90,208]. A nanoparticle containing both ¹⁸⁸Re and doxorubicin for the intravenous treatment of a colon cancer model has also shown a synergistic tumor treatment efficacy of this combination [209]. These studies demonstrate the significant promise that locally applied combination chemo-radionuclide carrying nanoparticles could be more effective in treating solid tumors by image-guided local application compared to each agent being applied separately. In glioblastoma therapy, ¹⁸⁶Re-liposomes could be combined with traditional intravenous glioblastoma chemotherapy with agents such as temozolimide and this has been proposed as a promising direction for future work [174].

6.7. Potential for immunostimulation of radiotherapeutic nanoparticles

In a recent study, intratumorally administered solid lipid nano-capsules carrying ¹⁸⁸Re have been shown to induce an anti-tumor immune response following local application in a rat brain tumor model [165]. This immune response has been attributed to the unique combination of the immunostimulatory properties of nanoparticles with the ability of ionizing radiation to induce immunogenic tumor cell death [165]. The authors of the paper have proposed that further strategies such as nanoparticle modification with immunostimulants could be used to increase an induced immune response against the tumor. Further investigations of the immune response induced by local nanoparticle radionuclide therapy appear promising. We agree and note that local beta-emission immunoradiotherapy has appealing potential in that the radiation therapy is mainly limited to the local tumor region, while appearing to have limited toxicity to the remaining immune system, thus providing potential of a significant immune response against anti-genic tumor tissue released in the local region following radionuclide nanoparticle therapy.

7. Conclusions

Image-guided radiotherapeutic nanoparticles have significant potential for solid tumor cancer therapy. The current success of this therapy in animals is most likely due to the improved accumulation, retention and dispersion of nanoparticles within solid tumor following image-guided therapies as well as the micro-field of the β-particle which reduces the requirement of perfectly homogeneous tumor coverage. It is also possible that the intratumoral distribution of nanoparticles may benefit from their uptake by intratumoral macrophages although more research is required to determine the importance of this aspect of intratumoral radionuclide nanoparticle therapy. This new approach to cancer therapy is a fertile ground for many new technological developments as well as for new understandings in the basic biology of cancer therapy. The clinical success of this approach will depend on progress in many areas of interdisciplinary research including imaging technology, nanoparticle technology, computer and robot assisted image-guided application of therapies, radiation physics and oncology. Close collaboration of a wide variety of scientists and physicians including chemists, nanotechnologists, drug delivery experts, radiation physicists, robotics and software experts, toxicologists, surgeons, imaging physicians, and oncologists will best facilitate the implementation of this novel approach to the treatment of cancer in the clinical environment. Image-guided nanoparticle therapies including those with β-emission radionuclide nanoparticles have excellent promise to significantly impact clinical cancer therapy and advance the field of drug delivery.