**4. Radiolabeled CD-Based Contrast Agents**

PET and SPECT are imaging modalities that require probe radiolabeling to produce tomographic images. These modalities have superior sensitivity and deeper tissue penetration compared with luminescence-based imaging and MRI [97]; they require lower dose of imaging agent for functioning. However, the application of PET/SPECT tracers is limited to the half-life of radiolabeled isotopes and biodistribution in the living organism.

The first demonstration of the CD-based PET molecular imaging probe was done by Bartlett et al., in 2007 [25]. CD-containing NPs were studied as delivery agents for transferrin (Tf)-targeted delivery to tumors of siRNA molecules. Nontargeted and Tf-targeted siRNA NPs were synthetized by using cyclodextrin-containing polycation. One percent of the targeted NPs containing adamantane-PEG molecules on the surface were modified with Tf. Si-RNA were conjugated with DOTA to the 5<sup>0</sup> end with further <sup>64</sup>Cu labeling [25]. MicroPET revealed negligible impact of the attachment of the Tf targeting ligand to the NPs on biodistribution. Unfortunately, nearly identical tumor localization kinetics of both targeted and nontargeted <sup>64</sup>Cu-DOTA-siRNA NPs were observed; tumor accumulation was also similar at 1 day after injection (≈1% ID/cm<sup>3</sup> ). However, bioluminescent imaging showed the intracellular localization and functional activity of siRNA delivered by Tf-targeted NPs in the tumor cells 1 day after injection.

Interestingly, the presence of CD in therapeutic compounds allows radio isotopic labelling for PET imaging of the drug circulation. In vivo biodistribution of the IT-101, clinically developed drug for cancer treatment, in mice with Neuro2A tumors was studied by Schluep et al. [98]. IT-101 contains molecule drug camptothecin (CPT) conjugated with β-CD based polymers (CDP), which acts as a carrier system for active molecule. The CDP site was labeled with <sup>64</sup>Cu through attaching of DOTA complex for microPET imaging; the obtained nanoparticle was similar to the IT-101 structure. Plasma pharmacokinetics of <sup>64</sup>Cu-IT-101 ere studied at 1, 4, and 24 h after injection with microPET/CT. It was shown that a low-molecular-weight fraction cleared rapidly through kidneys due to biphasic elimination profile, whereas remaining NPs circulated with a terminal half-life of ≈13 h. Biodistribution of IT-101 was examined at 24 h after administration; the highest tissue concentration was found in the tumor followed by the liver.

Another study conducted to investigate the effect of CD inclusion in NPs for drug delivery revealed the possibility of CD-based SPECT imaging agent [42]. Areses et al. compared the adhesive abilities and biodistribution of orally administered poly(anhydride) NPs and CD containing NP (CD-NP) in rats utilizing labeling with 99mTc for SPECT imaging. 99mTc-NP showed activity only in the gastrointestinal tract on SPECT images, whereas 99mTc-CD-NP revealed extended residence time in stomach: about 13% of 99mTc-CD-NP administered dose and 3% of 99mTc-NP given dose were found in the stomach after 8 h.

Liu et al., in 2011 [39], studied the improvement of the biodistribution of NPs using CD. Rare-earth UCNPs were modified by α-CD and OA for increasing of the water-solubility. UCNP-OA-CD complexes with Tm inclusion were labeled with <sup>18</sup>F (18F-UCNP(Tm)-OA-CD) for microPET imaging of ex vivo and in vivo biodistribution in mice at 5 min and 2 h. This was the first labeling of CD-based probe with an <sup>18</sup>F. Ex vivo imaging displayed rapid accumulation of NPs in the liver (~90.8% injected dose(ID)/g) and spleen (~62.5% ID/g) at 5 min and further decreasing of liver uptake to ~57.6% ID/g, while increased spleen accumulation to 118.9% ID/g after 2 h post-injection. In vivo microPET images were consistent with ex vivo biodistribution results and showed intense radioactive signals in the liver and spleen at 5 min after injection.

In addition to the previously discussed reports, a pre-targeted approach for molecular imaging probe development was presented by Hou et al. [99]. This work was based on biorthogonal conjugation chemistry between NPs, which have tendency to accumulate in tumors to enhanced permeability and retention (EPR) effect and radiolabeled imaging agents for PET imaging. One of NPs components was synthesized from CD-grafted polyethyleneimine (CD-PEI) and trans-cyclooctene N-hydroxysuccinimide (TCO-NHS), resulted in TCO/CD-PEI building block. Actual tumor-targeting NP (TCO⊂SNPs) was prepared via self-assembly from four different blocks TCO/CD-PEI, CD-PEI, adamantane-grafted polyamidoamine (Ad-PAMAM), and Ad-grafted polyethylene glycol (Ad-PEG) and injected to the tail vein of the mice with U87 glioblastoma cells. After EPR-driven accumulation in tumor, TCOresulted in TCO/CD-PEI building block. Actual tumor-targeting NP (TCO⊂SNPs can dynamically disassemble to release TCO/CD-PEI. After 24 h post-injection of TCO⊂SNPs, freshly prepared tetrazine compound radiolabeled with <sup>64</sup>Cu through DOTA (64Cu-Tz) was injected. Subsequently, distributed <sup>64</sup>Cu-Tz can undergo biorthogonal reaction with TCO/CD-PEI parts left after TCO⊂SNPs disassembling in vivo, yielding the dihydropyrazine conjugation adduct <sup>64</sup>Cu-DHP/CD-PEI, which acts as a contrast agent for tumors. Multiple microPET and anatomical CT images were acquired following the injection of pre-targeted NPs and <sup>64</sup>Cu-Tz compound with in vivo reaction (Figure 10a), along with two series of control PET imaging of a fully ex vivo prepared <sup>64</sup>Cu-DHP/CD-PEI adduct (Figure 10b) and free radiolabeled reporter <sup>64</sup>Cu-Tz (Figure 10c). Pre-targeted studies showed the accumulation and retention of radioactivity mainly in the glioblastoma tumor and liver and some nonspecific uptake by tissues. Supramolecular nanoparticles (SNP) control and probe ( <sup>64</sup>Cu-Tz) imaging did not present highly distinguishable tumor uptake. Although, high radioactivity was observed in the liver in all three cases, which can be explained by <sup>64</sup>Cu2<sup>+</sup> dissociation from DOTA ligand. This issue can be eliminated by using different radioisotopes for labelling.

Modification of CD by grafting alkyl chains (C6-C14) can lead to self-organization of obtained derivatives into NPs potentially useful for drug delivery [100]. Further co-nanoprecipitation of bio-esterified alkylated cyclodextrins with PEGylated phospholipids (PEG) can lead to surface-modified NPs [101]. Perret et al. researched the effect of the PEG chain length on the plasma protein absorptivity and blood kinetics of NPs [43]. β-CD derivatives with C10 alkyl chain (β-CD-C10) were co-nanoprecipitated with PEG with chain length of 2000 (125I-βCD-C10-PEG2000-NP) and 5000 Da ( <sup>125</sup>I-βCD-C10-PEG5000-NP) and radiolabeled with <sup>125</sup>I for SPECT ex vivo and in vivo biodistribution studies. In vivo SPECT/CT images were acquired at 10 min, 1, 3, 6, and 24h following the injection of NPs without PEG (125I-βCD-C10-NP) and with PEG (125I-βCD-C10-PEG2000-NP, <sup>125</sup>I-βCD-C10-PEG5000-NP). Hepatic activity was observed with all NPs; however, splanchnic activity was observed only with <sup>125</sup>I-βCD-C10-NP. Additionally, <sup>125</sup>I-βCD-C10-PEG5000-NP systems showed reduced elimination and increased circulating concentration following in vivo intravenous injection in comparison with other NPs.

β-CD-based rotaxane were used in developing theranostic shell-crosslinked NPs (SCNPs) by Yu et al. for improving drug delivery and controllable release in supramolecular medicine [102]. The core-shell-structured self-assembling NPs were obtained from polyrotaxanes consisted of amphiphilic diblock copolymer and the primary-amino-containing β-CD (β-CD-NH2), which undergoes complexation with poly(ε-caprolactone) (PCL) segment. In the gained structure, amphiphilic deblock copolymer acts as the axle, and β-CD-NH<sup>2</sup> acts as a wheel in complex with PCL, whose chains can experience hydrophobic interactions along with the perylene diimide (PDI) stoppers, which has a

tendency to π-π stacking interactions. Obtained SCNPs were labeled with radioactive <sup>64</sup>Cu through DOTA attachment to SCNPs (64Cu SCNPs@DOTA) for PET imaging of the dynamic biodistributions and accumulations of SCNPs in the main organs. HeLa tumor-bearing mice were imaged at various time-points after intra-venous injection 150 µCi of <sup>64</sup>Cu SCNPs@DOTA. Images revealed the high liver uptake of <sup>64</sup>Cu SCNPs@DOTA along with increasing of tumor uptake from the point of injection to 12 h post-injection with further start of clearance at 48 h. control PET imaging of a fully ex vivo prepared 64Cu-DHP/CD-PEI adduct (Figure 10b) and free radiolabeled reporter 64Cu-Tz (Figure 10c). Pre-targeted studies showed the accumulation and retention of radioactivity mainly in the glioblastoma tumor and liver and some nonspecific uptake by tissues. Supramolecular nanoparticles (SNP) control and probe (64Cu-Tz) imaging did not present highly distinguishable tumor uptake. Although, high radioactivity was observed in the liver in all three cases, which can be explained by 64Cu2+ dissociation from DOTA ligand. This issue can be eliminated by using different radioisotopes for labelling.

pre-targeted NPs and 64Cu-Tz compound with in vivo reaction (Figure 10a), along with two series of

*Molecules* **2020**, *25*, x FOR PEER REVIEW 15 of 27

polyethylene glycol (Ad-PEG) and injected to the tail vein of the mice with U87 glioblastoma cells. After EPR-driven accumulation in tumor, TCO⸦SNPs can dynamically disassemble to release TCO/CD-PEI. After 24 h post-injection of TCO⸦SNPs, freshly prepared tetrazine compound radiolabeled with 64Cu through DOTA (64Cu-Tz) was injected. Subsequently, distributed 64Cu-Tz can undergo biorthogonal reaction with TCO/CD-PEI parts left after TCO⸦SNPs disassembling in vivo, yielding the dihydropyrazine conjugation adduct 64Cu-DHP/CD-PEI, which acts as a contrast agent

**Figure 10.** Timeline of the injection protocol employed for (**a**) pre-targeted, (**b**) SNP control (64Cu-DHP⊂SNPs), and (**c**) free radiolabeled reporter (64Cu-Tz) studies. Representative in vivo **Figure 10.** Timeline of the injection protocol employed for (**a**) pre-targeted, (**b**) SNP control (64Cu-DHP⊂SNPs), and (**c**) free radiolabeled reporter (64Cu-Tz) studies. Representative in vivo microPET/CT images of the mice (n = 4/group) subjected to the three studies at 24 h p.i. Labels T, L, K, and B refer to the tumor, liver, kidney, and bladder, respectively. Dashed lines correspond to the transverse cross-section through the center of each tumor mass, whose image is shown in the right panel [99]. The images are reprinted with permission from publisher [99]. (**d**) The chemical structure of the tumor targeting imaging probe developed by Hou et al. [99].

Another study utilized PET for investigation in vivo distribution of 2-Hydroxypropyl-β-CD (HPBCD), β-cyclodextrin derivative, and an orphan drug for the Niemann–Pick disease treatment [40]. Six-deoxy-6-monoamino-(2-Hydroxypropyl)-β-CD (NH2-HPBCD) was conjugated with p-NCS-benzyl-NODA-GA (NODAGA) and radiolabeled with <sup>68</sup>Ga for PET/CT imaging. Ex vivo and in vivo studies on healthy mice showed that <sup>68</sup>Ga-NODAGA-HPBCD was mainly excreted through the urinary system with low uptake of the abdominal and thoracic organs and tissues at 30 and 90 min post-injection.

The most recent study in the area of CD-based PET molecular imaging probes containing tumor targeting compounds was done by Trencsenyi et al., in 2019 [41]. Their aims were to develop novel radiolabeled compound specific to the prostaglandin E2 (PGE2), which plays an important role in tumor progress and formation of metastases. The high affinity of PGE2 to the randomly methylated β-CD (RAMEB) was reported by Sauer et al. [103]. Trencsenyi et al., in their research, aimed to synthesize PGA-specific RAMEB labeled with <sup>68</sup>Ga through NODAGA (68Ga-NODAGA-RAMEB) for investigation of its tumor-targeting properties and in vivo biodistribution using PET. PancTu-1 and BxPC3 tumor-bearing SCID mice were intravenously injected with <sup>68</sup>Ga-NODAGA-RAMEB. The injection was followed with dynamic and static microPET imaging at 0–90 min. The accumulation of <sup>68</sup>Ga-NODAGA-RAMEB was significantly higher in BxPC3 tumors than in the PancTu-1; the highest post-injection tumor-background ratio (T/M) was obtained at 80–90 min post-injection. The T/M standardized uptake values (SUVs) were 10-fold lower in the PancTu-1 than those of BxPC3 tumors confirming the high PGE2 selectivity of <sup>68</sup>Ga-lebeled cyclodextrin.
