Nanoparticles labeled with Positron Emitting Nuclides: Advantages, Methods, and Applications

Abstract

Over the past decade, positron emitter labeled nanoparticles have been widely used in and substantially improved for a range of diagnostic biomedical research. However, given growing interest in personalized medicine and translational research, a major challenge in the field will be to develop disease specific nanoprobes with facile and robust radiolabeling strategies and that provide imaging stability, enhanced sensitivity for disease early stage detection, optimized in vivo pharmacokinetics for reduced non-specific organ uptake, and improved targeting for elevated efficacy. This review briefly summarizes the major applications of nanoparticles labeled with positron emitters for cardiovascular imaging, lung diagnosis and tumor theranostics.

INTRODUCTION

During the last decade, molecular imaging has expanded due to its unique suitability to support personalized medicine by using modified or engineered molecules that can reveal individual biology when coupled with an appropriate imaging approach. Molecular imaging approaches include both single modality such as positron emission tomography (PET), single photon emission computed tomography (SPECT), magnetic resonance imaging (MRI), magnetic resonance spectroscopy (MRS), computed tomography (CT), ultrasound, bioluminescence, and fluorescence imaging, and also multimodalities, such as PET/CT, SPECT/CT and PET/MR.¹ Among these approaches, the radionuclide-based imaging methods, especially PET, have been a particular focus in biomedical research due to advantages that include high sensitivity (picomolar level) and limitless tissue penetration.²

While many types of molecules have been used in molecular imaging, a growing area of interest is the use of nanoparticles, which have great potential for early detection, accurate diagnosis, and personalized therapy of various diseases, especially cancer.³ Nanoparticles are structures ranging in size from 1 to100 nm (Figure 1). Nanoparticles show unique size-dependent physical and chemical properties, which can be optical, magnetic, catalytic, thermodynamic, and electrochemical. ⁴ Generally, nanoparticles used for biomedical research can be categorized into three groups: 1) inorganic nanoparticles including quantum dots, iron oxide nanoparticles, gold nanostructures, and upconversion nanophosphors; 2) polymer nanoparticles such as core-shell dendrimers and amphiphilic nanoparticles; 3) lipid nanoparticles including liposomes and solid lipid nanoparticles. Additionally, radiolabeled carbon nanotubes and nanodiamonds have also been widely explored for oncological applications. ^5–8 Nanoparticles’ pharmacokinetics and biodistribution have been reviewed in detail elsewhere. ^{9, 10}

Figure 1.

Scheme of multifunctional nanoparticles

Nanoparticles’ design flexibility enables tunable in vivo pharmacokinetics to improve delivery efficacy and to reduce non-specific organ uptake by varying the size, charge, and surface modification. ¹¹ With a diameter about 100nm, nanoparticles show prolonged blood circulation and relatively low rate of mononuclear phagocyte system (MPS) uptake. ¹⁰ Their size also fills a critical position between the macroscopic world and molecular-level detail, and they can be designed to provide unique advantages over both macroscopic materials and molecular systems. Because their size is comparable to large biological molecules (antibodies, DNA), nanoparticles can be designed to interact with various biomolecules both on the surface and inside cells, leading to significantly improved diagnosis and treatment efficacy.³ Another noteworthy physicochemical characteristic is the nanoparticle’s high surface area to volume ratio, which enables rich surface chemistry for various targeting components while retaining high loading capacity for detection elements and therapeutic payload, as well as multifunctionality for synergistic applications (Figure 1).¹² Regarding charge, neutral nanoparticles are reported to demonstrate a slow clearance profile and reduced hepatic and splenic accumulations compared to charged nanoparticles,¹³ while positively charged nanoparticles administered intravenously often form aggregates due to the presence of negatively charged serum proteins.¹⁴ To reduce opsonization and aggregation caused by protein deposition, hydrophilic polymers such as polyethylene glycol are normally grafted onto the surface of nanoparticles, which also improves blood retention for optimized targeting and delivery. ¹⁵

Over the last decade, growth in the applications of nanoparticles in molecular imaging and drug delivery by utilizing gamma-emitting radioisotopes and positron emitters ^16–21 has led to drug discovery and numerous clinical trials. ^22–25 Due to broad application of PET isotopes in translational research, we focus on the major biomedical applications of nanoparticles radiolabeled with these positron emitters. ^26–30

APPLICATIONS OF PET RADIONUCLIDE LABELED NANOPARTICLES

PET radionuclide labeled nanoparticles have been extensively used in both preclinical and clinical studies as a tool to explore nanoparticles’ in vivo pharmacokinetics, imaging capability, and theranostic potential. ^{19, 22, 31, 32} For nanoparticles with different physicochemical properties and functional groups, the specific PET isotope and radiolabeling strategy need to be carefully considered to generate an optimal imaging outcome. The nuclear characteristics of commonly used PET radionuclides for nanoparticles are summarized in Table 1.

Table 1.

Nuclear characteristics of selected PET radionuclides for nanoparticles

Radionuclide	T1/2	Decay (%)	β Energy (KeV)		Main photon KeV (%)	Production
			Max.	Mean
68Ga	67.7min	β+ (89) EC (11)	1899	829	511 (178.3)	68Ge/68Ga generator
18F	109.7min	β+ (96.7) EC (0.1)	634	245	511 (193.5)	18O (p, n) 18F
64Cu	12.7h	β+ (17) EC (44)	653	278	511 (34.8)	64Ni (p, n) 64Cu
76Br	16.2h	β+ (55) EC (45)	3941	1180	511 (109); 559 (74) 657 (15.9); 1854	76Se (p, n) 76Br 76Se (d, 2n) 76Br
86Y	14.7h	β+ (33) EC (66)	3141	664	511 (63.9); 1077 (82.5)	86Sr (p, n) 86Y
89Zr	3.3d	β+ (23) EC(77)	901	397	909 (100)	89Y(p, n)89Zr
124I	4.18d	β+ (23) EC (77)	2138	820	511 (46); 603 (62.9) 723 (10.3)	124Te (p, n) 124I 124Te (d, 2n) 124I

There are two main radiolabeling strategies for nanoparticles. One is to radiolabel the nanoparticle structure itself, either on the surface or in the core. The other approach is to radiolabel the payload encapsulated inside the nanoparticle. These two approaches share much of their chemistry and are both widely used for nanoparticle radiolabeling (Table 2). ³³

Table 2.

Labeling strategies and specific activities of PET radionuclides labeled nanoparticles

Nanoparticle	Radionuclide	Labeling strategy	Specific activity*	Reference
Quantum dot	18F	nucleophilic substitution	3.7–7.5×108 Bq (10–20 mCi)/nmol	122
	64Cu	DOTA	3.7×107 Bq (1 mCi)/nmol	36
	64Cu	DO3A	6.2×105Bq (17μCi)/mg	123
Iron oxide	18F	Click chemistry	6.7±0.8×108 Bq (18±2 mCi)/mg Fe	73
	64Cu	DOTA	3.7–7.4×108 Bq (10–20 mCi)/mg Fe	26
	68Ga	Direct labeling	3.6×108Bq (10 mCi)/nM Fe	41
	68Ga	NOTA	1.5×108Bq (4 mCi)/nmol	124
Aluminum hydroxide	124I	Tyrosine	5.1×107 Bq (1.4 mCi)/mg (Fe+Mn)	42
	18F	Inorganic interaction	5.4×106Bq (146 μCi)/mg	125
Upconversion nanophosphors	18F	Inorganic interaction	7.8×108 Bq (21 mCi)/mg	126
Gold nanoparticle	64Cu	DOTA	5.9×1011 Bq (16 Ci)/nmol	44
Latex	68Ga	Direct labeling	2×105Bq (5 μCi)/mg	127
Liposome	64Cu	DOTA	13.3±1.0 ×105 Bq (36±3 μCi)/nmol	117
	64Cu	TETA, CB-TE2A	7.7±0.6 ×105 Bq (21±2 μCi)/nmol	128
	64Cu	BAT	2.1×107 Bq (0.6 mCi)/nmol	129
	18F	Encapsulation	2.8×107 Bq (0.8 mCi)/nmol	130
	18F	Encapsulation	1.1×105 Bq (3 μCi)/nmol	131
	68Ga	DTPA	4×106 Bq (0.1 mCi)/μg	132
Solid lipid nanoparticle	64Cu	BAT	1.4±0.3×106 Bq (38±8 μCi)/mg lipid	133
Polymer	76Br	Tyrosine	1.9×105 Bq (5 μCi)/μg	27
	64Cu	DOTA	1.5×107 Bq (0.4 mCi)/μg	63
	18F	[18F]FETos	30 Bq (0.8nCi)/μg	134
Nanotube	64Cu	DOTA	7.4–11.1×106 Bq (0.2–0.3 mCi)/μg	5
	89Zr	desferrioxamine B	592 KBq/μg	135
	86Y	DOTA	555 GBq/g	8

^*The data presented in Table 2 should be reviewed with caution. The values listed in the literature are specific activities quoted during nanoparticle radiolabeling using different units and have also been calculated using different analytical methodologies.

In designing a radiolabeled nanoparticle for biomedical applications, some key factors need to be considered. The first of these is radiolabeling integrity. For in vitro or in vivo applications of radiolabeled nanoparticles, the radionuclide itself is observed or detected rather than the nanoparticle or the payload. Thus the nanoparticle structure and radiolabeling strategy must both be designed to get robust, stable radiolabeling. The second factor to consider is application compatibility – the half-life of the radionuclide needs to be congruent with the binding kinetics of the probe and target, as well as the probe’s in vivo pharmacokinetics. A third factor is the targeting efficiency and radiolabeling specific activity, since a well-designed nanoparticle that allows increased loading of targeting ligands and high radiolabeling specific activity can provide elevated binding efficiency and reduce the required administration of nanoparticle to just trace amounts. The fourth factor is translational capability, since the U. S. Food and Drug Administration (FDA) approval for human application will be needed to explore the clinical potential of nanoparticles. For example, although there are chelators showing better ⁶⁴Cu radiolabeling stability for PET imaging,³⁴ DOTA (1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid) is still the most used chelator for translational research owing to its FDA approval and wide applications in clinical trials. ³⁵

A wide range of nanoparticles have been labeled with a variety of radionuclides. Quantum dots (QDs) are inorganic fluorescent semiconductor nanoparticles with desirable properties for optical imaging applications and have been radiolabeled with various radionuclides to explore their in vivo pharmacokinetics in an effort to develop multifunctional imaging probes. However, the hydrophobic nature of QDs leads to short blood circulation and insufficient targeting even after surface pegylation. ^{13, 36} More importantly, their potential toxicity limits the translational application of QDs.³⁷ Magnetic nanoparticles, especially iron oxide nanoparticles have also been widely explored for imaging applications because of high T₂ relaxivity for enhanced contrast, none radiation burden, biocompatibility and low clinical toxicity. ^{38, 39} In addition, although most applications of iron nanoparticles have been focused on developing MRI contrast agents, preparation of iron oxide based PET/MR dual functional nanoparticles has been an active research area. So far, various radiolabeling strategies with different positron emitters including ⁶⁴Cu, ⁶⁸Ga and ¹²⁴I have been used to study the in vivo biodistribution profile and targeting efficiency of iron oxide nanoparticles with high radiolabeling yields and specific activities in a variety of animal disease models. ^{26, 40–42} Another target is gold nanostructures, which have tunable optical properties in the near-infrared (NIR) region (650 to 900 nm) and are thus particularly attractive for hyperthermia based on the photothermal effect, leading to increased cancer theranostic applications. ^{9, 43} Of the available imaging modalities, PET is the most widely used technique to monitor the delivery of gold nanostructures due to its high sensitivity and quantitative detection. A recent study of a ⁶⁴Cu-radiolabeled gold nanoshell showed clear tumor uptake, indicating the potential for not only PET imaging but also as a theranostic agent. ⁴⁴

Polymer nanoparticles have been widely used for biomedical imaging applications using a variety of radiolabeling strategies due to the versatility of synthetic chemistry. The structural design and in vivo PET imaging of polymer nanoparticles is reviewed in detail elsewhere. ¹⁷ In addition, liposome nanoparticles have been used for drug delivery since their initial discovery 40 years ago and are available with a myriad of possible compositions and modifications. ⁴⁵ Significant progress has been achieved by utilizing liposomes as nanocarriers for both diagnosis and therapy. A wide variety of radionuclides and labeling strategies have been employed for generating radioactive liposomes, and these are reviewed elsewhere. ⁴⁶

Silica nanoparticles, due to the well-known biocompatibility, have also been explored for various biomedical applications with radiolabels. ^{47, 48} With ¹⁸F labeling, the thermally hydrocarbonized porous silicon nanoparticle demonstrated that the particles passed intact through the gastrointestinal tract after oral administration and were not absorbed from a subcutaneous deposit. With intravenous injection, a fast MPS clearance profile was confirmed. This silica nanoparticle exhibited excellent in vivo stability, low cytotoxicity, and non-immunogenic profiles, indicating the potential for oral drug delivery. ⁴⁷ In another study, an organically modified silica nanoparticle also showed no toxicity in vivo and full clearance through hepatobiliary excretion, which was confirmed by both ¹²⁴I and near infrared dye DY776 labeling. ⁴⁸ Lately, nanodiamonds have also been proposed as a promising biomaterial for drug delivery owning to the biocompatibility of this form of carbon. ⁴⁹ With ¹⁸F radiolabeling, these nanodiamonds showed high lung, liver and spleen uptake, and significant excretion through the urinary tract. ⁷ Another recently emerged nanostructure for oncological applications is known as an upconversion nanoparticle. This nanostructure has very fast radiofluorination kinetics and multimodality imaging properties, but its in vivo pharmacokinetics still need improvement to achieve sufficient blood circulation. ^50–52

With increasing support from the National Institute of Health to study and develop nanotechnology, ⁵³ additional applications of nanomedicine research is expected. Here we describe selected applications of radiolabeled nanoparticles with the focus on core-shell polymeric nanoparticles.

Radiolabeled nanoparticles for cardiovascular imaging

It is well known that systemically administered nanoparticles tend to be sequestered by the MPS system and to accumulate mainly in the liver and spleen. Clearance from the bloodstream depends on particle size, surface configuration, and several other factors, and the first step is opsonization that triggers complement activation and macrophage recognition.⁵⁴ To target the low abundance biomarkers in animal cardiovascular disease models, nanoparticles must have high radiolabeling specific activity and binding specificity and be able to circulate for a sufficient period of time in the bloodstream, which requires well-defined structure, composition, and controlled in vivo properties. Among various nanostructures (such as iron oxide, silica, and gold nanoparticles), ^55–59 amphiphilic core-shell nanostructures have received particular attention because of the tunable in vivo pharmacokinetics and versatile conjugation chemistry. ^{60, 61} Shell cross-linked knedel-like nanoparticles (SCKs) are comprised of a hydrophobic polystyrene core that can be used to load hydrophobic drug molecules, and a hydrophilic external shell of poly (acrylic acid-co-acrylamide) that provides additional sites for other functional units such as imaging moieties. Through various synthetic strategies, especially cross-linking, SCKs can be prepared with controlled size, surface charge, pegylation density, multi-functionality, and tuned in vivo pharmacokinetics. ^{54, 62} Additionally, the multivalency of SCKs empowers flexible radiochemistry (⁶⁴Cu, ⁷⁶Br, ¹²⁴I, and ¹⁸F) for PET applications. By conducting DOTA conjugation before nanoparticle assembly, the amount of DOTA accessible for ⁶⁴Cu labeling could be accurately controlled with more than 400 copies per SCK, leading to a specific activity greater than 1.48×10⁷ Bq/μg.⁶³

Another type of core-shell star-like or comb-like co-polymer could be prepared with nitroxide mediated living radical polymerization to create defined sizes and morphologies. In one example, the chelator DOTA was placed in an internal, hydrophilic environment allowing efficient ⁶⁴Cu radiolabeling to make a protected and high specific activity nanoscopic imaging probe. Biodistribution studies showed a distinct correlation between the length of PEG grafts and the in vivo circulation time; with increased PEG chain length, increased blood retention and reduced MPS system uptake were observed.⁶⁴ Furthermore, the cargo loading capacity of this type of nanoparticle can be adjusted while retaining similar physicochemical properties. In a recent study, varying amounts of RGD peptide (5%-50% RGD) were accurately conjugated to the shell of comb-like nanoparticles for targeting α_vβ₃ integrin, and these RGD-combs all maintained similar sizes and radiolabeling specific activities. ^{65, 66} The in vitro studies of the RGD-combs showed positive correlation between RGD peptide loading and uptake in α_vβ₃ integrin-positive U87MG glioblastoma cells, demonstrating the importance of controlled conjugation of targeting groups to achieve optimal targeting performance with multivalent nanoparticle systems. ⁶⁶ Further, the comb nanoparticles were conjugated with C-type atrial natriuretic factor (CANF) to target the natriuretic peptide clearance receptor (NPRC) in a mouse angiogenesis model. By controlling the number of DOTA conjugation, high specific activity (5.4±1.2GBq/nmol) of ⁶⁴Cu radiolabeling could be achieved, ensuring the trace administration of ⁶⁴Cu-DOTA-CANF-comb (7 pmol) for imaging studies. PET images showed significantly higher standardized uptake values (SUVs) at angiogenesis sites created by hindlimb ischemia compared to contralateral control sites. More importantly, the SUVs of ⁶⁴Cu-DOTA-CANF-comb were 3.4 times higher than those obtained with DOTA-CANF peptide tracer and about triple of those from the non-targeted ⁶⁴Cu-DOTA-comb, demonstrating the superiority of a multivalent nanoprobe over the corresponding monovalent CANF peptide for in vivo molecular targeting, (Figure 2).⁶⁷

Figure 2.

PET/CT images of ⁶⁴Cu-DOTA-CANF-Comb and ⁶⁴Cu-DOTA-Comb in the HLI induced angiogenesis model obtained 7 days after ischemia. (A) ⁶⁴Cu-DOTA-CANF-Comb in HLI model showing the accumulation of activity in the ischemic limb with little observed on the contralateral nonischemic limb. (B) ⁶⁴Cu-DOTA-Comb in HLI model showing the weak uptake in both ischemic and nonischemic limbs. (C) Uptake of ⁶⁴Cu-DOTA-CANF-Comb (n=8) and ⁶⁴Cu-DOTA-Comb (n=7). (D) Ischemic/nonischemic uptake ratios of ⁶⁴Cu-DOTA-CANF-Comb (n=8) and ⁶⁴Cu-DOTA-Comb (n=7). Figure reproduced with permission from reference 67.

In developing nanoparticles for targeted drug delivery, controlled release kinetics, bioavailability and reduced toxicity are key considerations, which have made biodegradable nanoparticles an active research area. ^68–70 Compared to the inorganic nanoparticles, the polymeric nanoparticles can be uniquely prepared with biodegradable core or crosslinker for programmed release of therapeutic payload through enzyme or pH response degradation (Figure 3), which greatly enhances their biocompatibility and makes them better candidates for targeted diagnosis and drug delivery. Thus, a core-shell biodegradable dendritic nanoprobe labeled with ⁷⁶Br has been prepared for targeting α_vβ₃ integrins expressed in a mouse angiogenesis model. The controlled introduction of targeting CRGDC peptide to the shell offered 50-fold enhancement of in vitro binding affinity to α_vβ₃ integrins relative to the monovalent RGD peptide alone. In vivo, specific targeting to α_vβ₃ was observed with the targeted nanoprobe demonstrating a 6-fold increase of receptor-mediated endocytosis at the injured site compared to the control nanoprobe (Figure 4).²⁷ Additionally, the potential of poly(lactide-co-glycolide) based biodegradable nanoparticles have also been assessed for PET imaging due to their FDA approval for human use. ⁷¹

Figure 3.

Diagram of the biodegradation process of shell crosslinked knedel-like nanoparticles

Figure 4.

Non-invasive PET/CT images of angiogenesis induced by hindlimb ischemia in a murine model. (A) Nontargeted dendritic nanoprobes (shown bottom center). (B) Uptake of α_vβ₃-targeted dendritic nanoprobes was higher in ischemic hindlimb (left side of image) as compared with control hindlimb (right side of image). Figure reproduced with permission from reference 27.

Iron oxide nanoparticles have been widely used in cardiovascular imaging with various radiolabels. With carbohydrates such as dextran coating and diethylenetriaminepentaacetic acid (DTPA) conjugation, the ⁶⁴Cu radiolabeled iron oxide nanoparticle (⁶⁴Cu-TNP) was used to target macrophage in an apolipoprotein E deficient (apoE^−/−) mouse model of aneurysms. The high specific activity (3.7×10⁸ Bq/mg Fe of nanoparticle) ensured lower dose administration (1.5 mg Fe/kg body weight) than that used in oncology clinical trials (2.6 mg Fe/kg body weight) and sensitive detection of nanoparticle accumulation in various organs. The in vivo biodistribution of ⁶⁴Cu-TNP showed sufficient blood circulation (t_1/2 ≥ 4h) and major accumulation in liver and intestine. PET/CT imaging clearly showed the significant localization of ⁶⁴Cu-TNP in the thoracic aorta with a target-to-background ratio of 5.1±0.9, indicating the clinical translatability of this radiolabeled nanoparticle. Furthermore, an ¹⁸F radiolabeled iron oxide nanoparticle (¹⁸F-CLIO) has been developed due to the wide availability, sensitivity, and covalent radiolabeling of this radioisotope.⁷² With rapid [¹⁸F] click fluorination, high radiolabeling efficiency and specific activity were achieved (6.8 ± 0.8 × 10⁸ Bq/mg Fe of nanoparticle). The in vivo pharmacokinetics studies showed comparable blood retention to the ⁶⁴Cu-TNP. In the apoE^−/− aneurysms mouse model, PET imaging showed that the avid internalization by phagocytic cells led to significantly higher tracer accumulation at aneurysms relative to wild-type aorta.⁷³

Radiolabeled nanoparticles for lung imaging

The incidence of respiratory disease and infections such as asthma, chronic obstructive pulmonary disease, cystic fibrosis, infectious disease, and tuberculosis is increasing worldwide. The classification of chronic respiratory diseases as a major disease burden by the World Health Organization has led to increased efforts to prevention, diagnosis and treatment of these diseases.⁷⁴ The current challenges for respiratory disease treatment include the sustained delivery and controlled release of drugs, reduction of side effects caused by high dose administration, and increasing drug resistance. Nanotechnology-based delivery systems have gained attention for use in pulmonary diagnosis and therapy due to their capacity for targeted deposition, bioadhesion, bioavailability, and biocompatibility, and their sustained release, which allows reduced dosing frequency and improves convenience for the patient.⁷⁵ So far, a variety of nanocarriers have been used for pulmonary applications including liposomes, solid lipid nanoparticles, metal nanoparticles, nanotubes, and polymeric nanoparticles.^{76, 77} Among these nanostructures, owing to the concern about the toxicity, those with potential clinical pulmonary applications such as polymeric nanoparticles, especially the ones made from biodegradable materials have been an active area in both lung diagnosis and treatment.⁷⁸

Recently, various materials such as poly(lactide-co-glycolide), polyacrylates, and polyacrylamide have been used for formulation of biodegradable nanoparticles. ^{75, 76, 79, 80} In contrast to the hydrophobic materials, the polyacrylamide based hydrogel offers excellent biocompatibility and hydrophilicity. It also is strongly endosome-disrupting, which makes it a candidate for the cytoplasmic delivery/imaging. ^{81, 82} With an acid-degradable crosslinker, the entrapped payload can be released in a pH-dependent manner inside endosomes. ⁸³ In addition, for better cellular delivery, a cell penetration peptide (CPP) can be used on the nanoparticles to get various cargos into the cells without disturbing the stability of the cell membrane and with low cytotoxic effects. ⁸⁴ In a recent study, it was reported that the optimal size for deposition in the deep lung for systemic delivery is approximately 1–3μm – microparticles rather than nanoparticles. ⁸⁵ Therefore, a nona-arginine functionalized polyacrylamide-based microparticle was synthesized to study the delivery efficiency of entrapped protein into non-phagocytic lung epithelial cells (BEAS-2B). In vitro results showed effective delivery of encapsulated BSA-Alexa Fluor 488 into the BEAS-2B cells in both CPP- and concentration-dependent manners, as well as a time dependency. ⁸⁶ As a result, this CPP-modified microparticle was labeled with radiohalogens (¹²⁵I and ⁷⁶Br) for animal studies to assess the in vivo fate, lung retention, and cellular uptake after intratracheal administration. Furthermore, nanosized CPP particles were also synthesized to compare size-related differences in the clearance profiles. The biodistribution studies revealed that particle retention and extrapulmonary distribution was, in part, size dependent. Microparticles were rapidly cleared by mucociliary routes but, unexpectedly, also through circulation. In contrast, nanoparticles had prolonged lung retention enhanced by the CPP, which was confirmed by the PET imaging analysis with ⁷⁶Br-radiolabeled nanoparticles (Figure 5). The studies indicate the potential of microparticles for short-term applications and benefits of nanoparticles for serial imaging or therapy of a persistent lung injury. ²⁹ In contrast, a study of acute lung injury used latex nanoparticles coated with anti-intercellular adhesion molecule-1(ICAM-1) antibody and labeled with ⁶⁴Cu for targeting the pulmonary endothelium. Biodistribution studies showed 3- to 4-fold higher uptake in the lungs of mice injected with ICAM-targeted nanoparticles than those receiving control nanoparticles. PET imaging clearly demonstrated the accumulation of radioactivity in the lungs. However, metabolic studies showed that the in vivo stability of this nanoprobe needs further improvement for prolonged pulmonary drug delivery. ²⁸

Figure 5.

Three dimensional reconstruction of microPET/CT imaging of ⁷⁶Br-labeled particles in mice lungs following intratracheal delivery. Arrows indicate gastrointestinal tract activity. Fiduciaries (f) used for coregistration are included. (A) microparticles; (B) nanoparticles. Figure reproduced with permission from reference 29.

CPP−: without cell penetration peptide: CPP+: with cell penetration peptide

Lately, a new type of promising biomaterial-carbon known as a nanodiamond has been explored for biomedical applications due to its biocompatibility, ability to cross the cell membrane, and capability to be functionalized to act as carriers. The initial biodistribution and PET imaging via ¹⁸F radiolabeling showed high pulmonary retention, most likely by size exclusion, indicating potential for lung applications. ⁸⁷ Clinically, ⁶⁸Ga-labeled carbon nanoparticles have also been used for pulmonary embolism PET/CT ventilation-perfusion imaging and has demonstrated its superiority to conventional V/Q lung scintigraphy. ⁸⁸

Radiolabeled nanoparticles for tumor imaging

One hundred years ago, Paul Ehrlich proposed the idea of “magic bullet” for the development of medicine to specifically target the cancer disease. ⁸⁹ Recently, development of molecular biology and genetic research, two major antitumor strategies have been revealed: 1) utilization of molecularly targeted therapeutics to block hallmarks of cancer, and 2) development of novel drug delivery systems utilizing tumor-specific nanomedicines to improve the pharmacokinetics and bioavailability of vehicle-carried drugs. ⁹⁰ Because of the versatile physiochemical properties of the nanostructure, in contrast to conventional anti-cancer drugs, nanoparticles can provide significant improvements in pharmacokinetics, targeting specificity and efficiency, diagnostic and therapeutic efficacy, and toxicity, which could lead to earlier detection and better control of cancer. ⁹¹ In development of nanoparticle-based agents for cancer diagnosis and therapy, important factors are active targeting of biomarkers expressed in the tumor and harnessing the enhanced permeability and retention (EPR) effect due to the leaky neovasculature of the tumor proposed by Matsumura and Maeda ⁹² that can be used for “passive targeting”. ⁹³ Compared with conventional small molecule-based anticancer drugs, macromolecules display superior in vivo pharmacokinetics and greater tumor delivery and selectivity. ^{23, 93} Interestingly, a size relationship with the EPR effect was observed that larger and long-circulating macromolecules (> 30–45 kDa) are retained in the tumor tissue longer, whereas smaller molecules easily diffuse back out into the bloodstream. ⁹² Nanoparticle size, surface modification, and vascular mediators all have been studied as approaches to harness the EPR effect for improved tumor diagnosis and therapy using nanoparticles. ^{93, 94}

While the EPR effect is helpful, nanoparticles also offer the ability to specifically target tumors based on an individual patient’s biology by targeting various biomarkers on the cancer cells. To date, a range of targeting groups have been developed and used for cancer nanomedicine. ^{3, 95–99} Among them, antisense-based imaging agents such as phosphodiester (PO) oligodeoxynucleotides (ODN) and phosphorothioate (PS)-ODNs, which are designed according to the gene expression profile of human cancerous cells, are promising imaging probes for the early and specific detection of cancer due to the high specificity. ¹⁰⁰ However, the rapid degradation by endo- and exonucleases in vivo made their PET imaging challenging. Alternatively, with the complete replacement of the sugar-phosphate backbone to amine backbone (peptide nucleic acids or PNA) through chemical modification, the PNAs displayed strong resistance to enzymatic degradation without changing the binding affinity and specificity. Therefore, a specially designed PNA sequence was used to target the elevated expression of upstream of N-ras (unr) mRNA in a mouse MCF-7 breast tumor model with ⁶⁴Cu radiolabeling. PET imaging clearly showed the tumor accumulation of ⁶⁴Cu-DOTA-PNA50-K4, indicating the potential of this antisense PNA as a specific molecular probe for cancer diagnosis. ¹⁰¹ Thus, this antisense PNA (PNA50) was conjugated to well-defined SCK nanoparticles for further evaluation. The in vivo pharmacokinetic studies demonstrated improved biodistribution profiles relative to PNA alone tracer while maintaining the PNA binding capability to target, indicating the potential of this nanoprobe for sensitive and specific cancer diagnosis. ¹⁰²

In another study, SCKs conjugated with folate showed specific interaction with folate receptors overexpressed in KB cells. In vivo studies with ⁶⁴Cu labeling demonstrated improved blood retention and folate receptor-mediated uptake of SCKs in small tumors. ¹⁰³ To develop sensitive nanoprobes for cancer diagnosis, the click chemistry strategy has also been explored on SCKs to obtain ultrahigh specific activity ⁶⁴Cu/¹⁸F radiolabeled nanoparticles as well as controlled conjugation of targeting ligands by designing click sites both in the core and on the surface. ^{104, 105}

Iron oxide nanoparticles have been widely explored for PET/MR or PET/MR/optical tumor imaging. ^106–108 With optimized surface pegylation and DOTA functionalization, a ⁶⁴Cu radiolabeled iron oxide nanoparticle (specific activity = 3.7–7.4×10⁸ Bq/mg Fe) showed elevated blood retention of 37.3± 12.9 %ID/g at 1h post injection in mice, confirmed by PET imaging (Figure 6). In another study, a cyclic RGD peptide (c(RGDfC)) was conjugated to a superparamagnetic iron oxide nanoparticle (SPIO) for targeted PET/MR tumor imaging. ^{109, 110} This ⁶⁴Cu-cRGD-SPIO illustrated low (<15 %ID/g) hepatic burden up to 48h and constant tumor uptake (~5% ID/g) across the study with the highest (11.3±2.5) tumor/muscle ratio observed at 48h. Furthermore, compared to the control ⁶⁴Cu-SPIO, ⁶⁴Cu-cRGD-SPIO had significantly (p<0.05) higher tumor accumulation during the study, indicating α_vβ₃-specific targeting.⁴⁰

Figure 6.

Coregistered microPET/microCT of BALB/c mice administered 100 μCi of ⁶⁴Cu-mSPIOs (10 mg Fe/kg body weight, 100 μL injection volume). Whole body sagittal (A-E) and coronal (B-F) PET images are decay corrected and scaled by min/max frame: (A, B) 1 h; (C, D) 4 h; and (E, F) 24 h post injection. Figure reproduced with permission from reference 26.

Lately, Cerenkov luminescence imaging has become an active topic in biomedical research due to the combination of nuclear tomography with optical techniques generated from the decay of radionuclides, and suitability for the rapid, high-throughput screening. ¹¹¹ Thus, a ¹²⁴I radiolabeled iron oxide nanoparticle was developed for optical/PET/MR tri-modality tumor imaging. The complementary nature of this hybrid nanoprobe facilitated non-invasive differentiation between tumor-metastasized sentinel lymph nodes (SLNs) and tumor-free SLNs.¹¹²

Silica nanoparticles, which are inherently nontoxic and biocompatible, have been an attractive candidate for theranostics in various patient settings. With cRGDY conjugation, high in vitro binding affinity (IC₅₀ = 1.2 nM) was achieved. In animal melanoma model, this targeted ¹²⁴I-cRGDY-PEG-dots nanoprobe showed optimized pharmacokinetics (blood and tumor half-lives = 5.9 hours and 73.5 hours, respectively), α_vβ₃-specific tumor uptake, and high tumor-to-muscle ratio (T/M=5 at 24h). Dosimetry studies demonstrated comparable radiation doses to other clinically used PET tracers. Toxicity studies confirmed full clearance in one week and no tissue-specific pathologic effect. Therefore, a human clinical trial has been planned to investigate the potential of this targeted nanoprobe in staging metastatic disease in the clinical setting. ²⁴

Another effort to develop theranostic agents involves use of various gold nanostructures such as nanoparticles, nanorods, nanoshells, and nanocages due to the well-known biocompatibility. They can serve as optical imaging agents due to the surface plasmon resonance, or PET imaging probe through surface conjugation. More importantly, their photothermal properties empower the conversion of absorbed light into heat through nonradiative electron relaxation dynamics for cancer treatment. ^{9, 43} Thus, gold nanoshell was used for multimodality theranostics with ⁶⁴Cu radiolabeling and RGD peptide conjugation. PET/CT of this ⁶⁴Cu-NS-RGDfK showed significant tumor uptake and tumor vascular specificity, indicating the active targeting and improved efficacy of photothermal ablation. ¹¹³ Additionally, a chelator-free [⁶⁴Cu]CuS nanoparticle with controlled specific activity was also prepared for tumor theranostic application. The pegylated [⁶⁴Cu]CuS nanoparticles showed about 15% ID/g blood retention at 4h in biodistribution studies and high tumor-to-muscle ratio (T/M=6.55) at 24h. Interestingly, this PEG-[⁶⁴Cu]CuS nanoparticle also displayed the photothermal property. ¹¹⁴

Liposomes, widely used for drug delivery in both clinical and pre-clinical applications, ¹¹⁵ can also be a good theranostic candidate. ¹¹⁶ Initially, liposomes with ⁶⁴Cu radiolabeling have been used to probe the EPR effect in tumor-bearing mice. With a remote loading approach for ⁶⁴Cu encapsulation, improved radiolabeling stability and tumor accumulation ((5.0 ± 2.0% ID/organ) was obtained, confirmed by PET/CT. ¹¹⁷ This approach was used to study the EPR effect during the transition from premalignant to malignant cancer in a mouse ductal carcinoma model. With disease progression, the vascular volume fraction increased 1.6-fold and the apparent vascular permeability to liposomes increased about 2.5-fold. Thanks to the long in vivo half-life (t _1/2=18h), high tumor/muscle ratio (17.9±8.1) was achieved with ⁶⁴Cu-liposomes. Interestingly, more heterogeneous intratumoral distribution was observed in the presence of increased vascular permeability. ¹¹⁸ More importantly, by adding encapsulation of EGFR kinase-targeting group SKI212243, this targeted liposome showed significantly higher tumor accumulation at 48h relative to ¹²⁴I-SKI212243 alone and greatly improved tumor-to-background contrast ratio at 48h, indicating specific targeting (not just the EPR effect due to improved circulation). ¹¹⁹

Lately, with improvement of nanoparticle in vivo pharmacokinetics, use of ⁸⁹Zr for nanoparticle PET oncological imaging has gained interest owing to its noteworthy physical properties, including long half-life (t_1/2=78.4h) and high specific activity. ¹²⁰ In an LS174T colon carcinoma model, the ⁸⁹Zr-labeled single wall carbon nanotubes (SWCNT-([⁸⁹Zr]DFO)(E4G10) showed rapid tumor accumulation and gradually increased tumor-to-muscle contrast ratio over time (1.61 at 1h to 5.08 at 96h). In another colon carcinoma model (CT26), an ⁸⁹Zr-labeled cross-linked dextran nanoparticle showed primary localization in lymph node (34 ± 16%ID/g). In some tumor bearing mice, PET imaging showed intense tumor uptake (20 ± 5%ID/g), surprisingly higher than other MPS organs, indicating translational potential. ¹²¹

CONCLUSION

A variety of nanoparticles has been engineered and explored for diagnostic and therapeutic potential in various diseases. The examples presented in this review focus on nanoparticles labeled with PET isotopes for cardiovascular, pulmonary and tumor imaging, as well as for pharmacokinetic evaluation. So far, significant progress has been achieved in nanoparticle structure design, in vitro trafficking, and in vivo fate mapping by using PET. More effort will be necessary to achieve active targeting and quantification of low level biomarkers expressed in animal models using customized nanoparticles generated through new chemistry for early disease detection and prevention with PET, and to achieve development of approved biocompatible and biodegradable nanoparticles for personalized medicine and translational research.