2. Liposomes
Liposomes are nanosized vesicles consisting of a lipid bilayer that can also contain cholesterol. They have an aqueous core and can be filled with either hydrophobic drugs (encapsulated in the bilayer) or hydrophilic drugs (encapsulated in the aqueous core). By means of a lipid chain, the main molecules (e.g., drugs and targeting agents) can also be linked to the membrane surface during liposome manufacture or via post-synthesis
[6]. Embedding drugs in liposomes improves their properties, achieving a better biodistribution and a lower toxicity
[7], and that is why these liposomal vesicles are often used for conventional drug delivery.
Liposomes were found to accumulate at tumor sites thanks to the enhanced permeability and retention (EPR) effect. They are easily taken up by the reticuloendothelial system (RES), however
[7], consequently accumulating in organs such as the liver and spleen. To avoid their rapid clearance, polyethylene glycol (PEG) chains can be conjugated to the liposome surface to extend their circulation time and enhance their accumulation at tumor sites
[7].
Studies on liposomes as drug delivery agents for use with radionuclides began in the last decade and led to the synthesis of a liposomal imaging tool called Vescan, which was never commercialized as it proved unable to detect tumors
[8][9].
Attention has now shifted from the radiolabeling of empty liposomes to the radiolabeling of liposomal formulations of conventional chemotherapeutics. The aims of these contents on this topic are to identify the pharmacokinetic (PK) properties of liposomal formulations once injected in vivo, and to establish which patients will better respond to this therapy.
Although liposomes are not a perfect example of a TRT, examining progress made in research on these vesicles can probably help people to better understand their potential future uses. Here are 16 articles on the radiolabeling of liposomal formulations of conventional chemotherapeutics: nine studies dealing with the pharmacokinetics of different liposomal formulations, and seven studies dealing with the patients’ different responses to the administration of liposomal formulations of chemotherapeutics.
The pharmacokinetic properties of liposomes may be influenced by the presence of a targeting agent on their surface. To give an example, Du and coworkers
[10] synthesized liposomes functionalized with MoAbs against programmed cell death-1 (PD-1), a receptor selectively expressed in triple-negative breast cancer. These liposomes were then filled with doxorubicin (DOX) and dual-labeled with a fluorophore (IRDye800WC) and a radionuclide (
64Cu). They proved better able to target and to treat the tumor due to the simultaneous effect of the MoAbs against PD-1 both as a targeting agent for liposomes and as an adjuvant immunotherapy for doxorubicin.
Alongside the presence of the targeting agent, the composition of a liposome may also influence its pharmacokinetic properties. Silva and coworkers
[11] demonstrated that long-circulating, pH-sensitive liposomes (SpHL) containing [
99mTc] DOX accumulated more in the tumor and were less active in the spleen and liver than liposomes that were not pH-sensitive. That said, Monteiro and coworkers
[12] noted that the presence of folate on the surface of SpHL (filled with paclitaxel) may lead to an even more sustained and higher tumor-to-muscle ratio than in the case of nonfunctionalized liposomes.
In addition to the presence of a targeting agent, other physical characteristics may enhance liposome delivery. For instance, Yang and coworkers
[7] were able to obtain a good tumor brain delivery of their liposomal formulation of DOX, with a high tumor-to-contralateral brain ratio. They associated the presence of a targeting agent (AP-1, a peptide capable of binding IL-4 receptor) with the focused ultrasound technique, which enables a temporarily disruption of the blood–brain barrier. Reversible electroporation may also enhance delivery to the tumor, with or without any targeting agent on the liposome’s surface; this technique enhances vascular permeability, altering the EPR effect and, thus, leading to a greater liposome deposition at the tumor site
[13].
To better study liposome distribution, the fluorescence technique can be associated with imaging, using positron emission tomography (PET), as in the earlier-mentioned work by Du et al.
[10]. Li and coworkers
[6] also succeeded in developing liposomes suitable for this application; their formulation could be labeled with the fluorophore IRDye-DSPE and the radionuclides
99mTc,
186/188Re, or
64Cu thanks to the presence of DOTA on the liposome’s surface
[6].
Luo and coworkers
[14] demonstrated that adding porphyrin phospholipid to the liposome’s bilayer may also be useful for the development of liposomal vesicles suitable for multimodality imaging.
Double radiolabeling is another way to obtain more information about the final target of both the liposome and the encapsulated drug. The feasibility of this technique was demonstrated by Lamichhane and coworkers
[15], who labeled the liposome’s surface with
111In and the carboplatin derivative it encapsulated with
18F. More attention has also been paid in recent times to the search for new radiotracers compatible with the half-life of liposomes, and
52Mn has been identified as a suitable radionuclide for this purpose
[16].
Pharmacokinetic studies have revealed a marked variability in liposome uptake by different tumors. This may be linked to the tumor’s mass, as Lin and coworkers
[17] found in their study; small tumors showed growth inhibition with all the treatment regimens tested (liposomes containing chemotherapeutics and/or radionuclides), whereas the growth of large tumors was only significantly inhibited by a combination of chemo- and radiotherapy. It is not unusual to see a different liposome uptake in different patients with the same tumor or different tumors in the same patient. Tumor deposition is due mainly to the EPR effect, which could complicate pretreatment planning and hamper predictions regarding a patient’s prognosis
[18].
The abovementioned studies on the pharmacokinetic properties of radiolabeled liposomes enabled tumor deposition and distribution to be quantified
[19], making it possible to identify patients mostly likely to respond to a liposomal therapy. Before testing liposomes in humans, it was important to demonstrate the feasibility of radiolabeling preformed liposomal formulations. This was the goal of a study by Edmonds and coworkers
[18], who successfully labeled liposomal formulations of drugs containing metal-binding motifs (e.g., doxorubicin and alendronate) with PET isotopes (e.g.,
89Zr,
52Mn, and
64Cu) using metal ionophores (e.g., hydroxyquinoline).
The uptake of liposomal formulations can also be studied by recreating liposomes with the same lipid composition. This was achieved in vivo by Ito and coworkers
[20], who synthesized liposomes with the same lipid composition as Doxil (a liposomal formulation of doxorubicin); they found a correlation between the therapeutic effect of Doxil and a histological factor associated with the EPR effect.
Clinical studies on the biodistribution of liposomal formulations of chemotherapeutics in patients were made by Arietta and coworkers
[21][22] and Lee and coworkers
[23]. Arietta’s group examined the antitumor activity of a therapy combining liposomal doxorubicin (LD) with cisplatin in patients with malignant pleural mesothelioma. They labeled the LD with
99mTc and found that patients who showed a
99mTc-LD uptake of 75% or more had significantly better rates of response, progression-free survival, and overall survival than patients with uptake levels below 75%. The authors concluded that
99mTc-LD uptake could be an important biomarker for use in assessing the results of therapy with LD and cisplatin
[21][22]. Lee’s group radiolabeled MM-302, an HER2-targeted Doxil formulation, with
64Cu. After promising preliminary in vitro results
[19], the liposomal vesicles were administered in humans
[23], and the
64Cu-MM-302 uptake was found to vary considerably, both across multiple lesions in the same patient and across different patients. A high uptake in the liver was due to the physiological metabolism of liposomes.
3. Avidin–Biotin Interaction
Avidin is a 66 kDa highly glycosylated, positively charged protein (isoelectric point~10) derived from egg white. It is tetrameric, and each monomer has a strong affinity for biotin (K
d = 10
−15)
[24][25][26].
The strength of the avidin–biotin interaction is such that it is considered irreversible, and this explains why its applications have been the object of so much interest. For example, it has been studied in the sphere of tumor-targeted therapy for use in a pretargeting approach, which consists of delivering MoAbs and radionuclides separately. The radionuclide delivery is delayed until the MoAbs have reached the maximum tumor-to-normal tissue ratio
[27], and the avidin–biotin interaction ensures the binding of the radiolabeled agent to the previously delivered antibody
[28]. Two- or three-step protocols have been used in this setting (
Figure 1).
Figure 1. Scheme of the two-step (A) and three-step (B) protocols.
In the two-step protocol, the tumor is first targeted with cold biotinylated antitumor MoAbs, and then radioactive-labeled avidin is administered
[27]. The three-step protocol involves (1) tumor pretargeting with cold biotinylated antitumor antibodies, (2) administering cold avidin to remove circulating biotinylated antibodies and ensure avidination of the biotinylated tumor-bound antibodies, and (3) labeling the tumor with radioactive biotin derivatives
[28]. Both protocols have been tested in preclinical and clinical settings, but only the clinical studies are considered. Here are nine articles that could be divided according to the type of tumor treated: three articles concerning various types of tumors, three articles concerning gliomas, and three articles dealing with breast cancer.
Paganelli and coworkers
[28] first tested the feasibility of the three-step protocol in 20 patients with tumors expressing carcinoembryonic antigen (CEA). Their preliminary study, conducted with
111In, revealed the advantages and disadvantages of the technique compared with the direct administration of radiolabeled MoAbs. The advantages included a drastically reduced background radioactivity, a well-preserved MoAb immunoreactivity (as autoradiolysis-induced damage to the MoAbs was avoided), and signal amplification. The main disadvantages were the need for repeated injections and the immunogenicity of avidin.
Some years later, Cremonesi and coworkers
[29] described the pharmacokinetic properties of the three-step protocol. The organs receiving the highest doses of radioactivity were the kidneys, liver, and urinary bladder, but the levels of renal, hepatic, or hematological toxicity were low.
Paganelli and coworkers
[27] also examined the application of the two-step protocol in the treatment of 15 patients with ovarian carcinoma. Here again, the high tumor-to-normal tissue ratio was highlighted as the main advantage of the method, and the repeated injections and use of streptavidin (an avidin analog) were identified as the main drawbacks.
After these first promising reports, the use of pretargeting strategies in tumor therapy spread and came to be applied to the treatment of malignant high-grade gliomas. In a phase I/II study, Paganelli and coworkers
[30] used a three-step protocol to deliver 15 times more radioactivity to the sites of brain tumors than to critical organs (e.g., liver and kidneys). The treatment’s toxicity was consequently acceptable, with most of the activity not bound to the tumor eliminated in the first 24 h. The therapeutic benefit was evident in most patients (the tumor progressed no further in 52% of cases and shrank significantly in 25%), and the response persisted for more than 1 year in some patients. In their study, the only major drawback was again immunogenicity due to the administration of streptavidin. To solve this problem, the authors recommended using avidin modified with polyethylene glycol (PEG) molecules instead of streptavidin, as PEG can hide avidin from the immune system. Grana and coworkers
[31] also tested a three-step protocol in the treatment of malignant gliomas, with promising results. The authors suggested that associating this technique with surgery, radiotherapy, and chemotherapy might increase the life expectancy of patients with high-grade gliomas. When Paganelli and coworkers
[32] applied the three-step protocol in the locoregional treatment of patients with high-grade gliomas, they reported an objective therapeutic response in many patients, along with an encouraging median overall survival. On the basis of the neurological toxicity observed, they identified 1.11 GBq as the maximum tolerated dose.
All these studies on the avidin–biotin interaction led Paganelli and coworkers to develop a new procedure called IART (intraoperative avidination for radionuclide therapy) for use in the treatment of breast cancer. This procedure consists of two main steps: (1) “avidination” of the anatomical area of the lesion directly after tumor resection; (2) intravenous injection of radiolabeled biotin to target the anatomical area of the tumor 1 day after surgery. Before the radiolabeled biotin is injected, the circulating avidin is removed by injecting an appropriate amount (20 mg) of biotinylated albumin
[24][25][33]. The rationale behind this procedure is that the inflammatory reaction after surgery makes the breast tissue a cation exchanger, thus enabling avidin retention at the site affected for several days
[24][25][33]. The first studies using IART generated information on the biodistribution of biotin, which was labeled with
111In via the DOTA chelator. The radiolabeled biotin uptake appeared to be fast and stable at the operated tumor site, with a rapid blood and renal clearance, as well as a consequently reduced toxicity. The doses absorbed by the most affected organs (bladder and kidneys) were well below the threshold doses reported in the literature
[24][33]. A more recent, phase II study was performed by Paganelli and coworkers
[25] to quantify the doses administered with IART. The biologically effective dose (BED) to the tumor bed was 21 Gy when a fixed activity of 3.7 GBq of
90Y-DOTA-biotin was injected. The authors judged that IART can consequently be considered as a boost to tumor treatment, especially in association with EBRT. They concluded that this technique may be applicable not only to any breast cancer amenable to conservative surgery, as well as to many other solid tumors such as those involving the bladder, prostate, and brain
[24][25][33].
4. Docking
Another possible strategy for performing an accurate TRT is to use a ligand with a very high affinity for the target. To ensure the strongest and most specific interaction, the ligand can be designed ad hoc, according to the structure of other known ligands or to the ligand’s interaction with the receptor. One way to ascertain whether the ligand thus designed is active is to use docking, a virtual simulation of ligand–receptor binding. The simulation of the interaction returns a score that can be used to quantify the ligand’s ability to bind its receptor.
Here are five recent studies on the application of docking in the development of a ligand suitable for radionuclide delivery. In three of the five articles, docking was used to select the ligands with the strongest interaction with the receptor, while it was used to justify the results obtained in the other two articles. Some studies dealt with imaging rather than therapeutic goals (as the use of docking in ligand development is relatively new, there has been too little time to assess all the ligands for therapeutic applications), but introduce them here to better explain the docking method.
Yang and coworkers
[34] designed several new ligands for prostate-specific membrane antigen (PSMA) and used docking to identify the most promising among them. They synthesized two series of ligands based on a carbamate structure; one contained the amino-pentanedioic acid (NPA) moiety, while the other contained the oxypentanedioic acid (OPA) moiety. Then, they used docking to test the interaction between the carbamate derivatives and the PSMA. Their results showed that the Lys-OPA carbamates were better ligands than the Lys-NPA carbamates. Two of the former showed a high target-selective uptake in tumor xenografts, and one of the two (4-bromo-2-[
18F] fluorobenzoyllysine OPA carbamate) also had a rapid normal organ clearance, making it the most likely candidate for clinical application.
Somatostatin receptor 2 (SSTR2) was the object of efforts to develop new ligands for treating neuroendocrine tumors (NETs). This receptor is the target of the previously mentioned somatostatin analogs DOTA-TOC and DOTA-TATE, but the focus of attention has recently shifted to developing receptor antagonists because they can bind to a larger number of sites, leading to a higher tumor uptake
[35]. With this in mind, Behnammanesh and coworkers
[35] developed a series of SSTR2 antagonists and labeled them with
177Lu by means of the DOTA chelator. Docking analysis then helped the authors to identify the peptide with the most successful accommodation at the binding site of the receptor (the DOTA-peptide 2, DOTA-p-Cl-Phe-Cyclo(
d-Cys-
l-BzThi-
d-Aph-Lys-Thr-Cys)-
d-Tyr-NH
2). The same peptide was subsequently synthesized and tested in vitro and in vivo; it showed a good stability, had suitable pharmacokinetic properties, and was able to reveal tumor lesions, making it a promising therapeutic agent for NETs.