Radiolabeled PSMA Inhibitors

Simple Summary

Prostate cancer remains one of the leading causes of cancer death in men worldwide. Despite the recent success in the development and clinical application of radiopharmaceuticals targeting the prostate-specific membrane antigen (PSMA) for diagnosis and endoradiotherapy of prostate cancer, more research is ongoing to further investigate and improve patient care and quality of life. Herein, an overview of novel developments and applications for small molecule and low-molecular weight radiolabeled PSMA inhibitors with an outlook to clinical translation is given.

Abstract

PSMA has shown to be a promising target for diagnosis and therapy (theranostics) of prostate cancer. We have reviewed developments in the field of radio- and fluorescence-guided surgery and targeted photodynamic therapy as well as multitargeting PSMA inhibitors also addressing albumin, GRPr and integrin α_vβ₃. An overview of the regulatory status of PSMA-targeting radiopharmaceuticals in the USA and Europe is also provided. Technical and quality aspects of PSMA-targeting radiopharmaceuticals are described and new emerging radiolabeling strategies are discussed. Furthermore, insights are given into the production, application and potential of alternatives beyond the commonly used radionuclides for radiolabeling PSMA inhibitors. An additional refinement of radiopharmaceuticals is required in order to further improve dose-limiting factors, such as nephrotoxicity and salivary gland uptake during endoradiotherapy. The improvement of patient treatment achieved by the advantageous combination of radionuclide therapy with alternative therapies is also a special focus of this review.

1. Introduction

Radiopharmaceuticals targeting the prostate-specific membrane antigen (PSMA) have seen great success in Nuclear Medicine for both diagnosis and endoradiotherapy of prostate cancer (PCa) [1,2] thus paving the way for (radio)theranostics [3,4,5] with guidelines being published by international experts and societies [6,7,8,9,10,11,12,13,14,15,16,17,18,19].

[⁶⁸Ga]Ga-PSMA-11 (gozetotide) [20,21,22,23] and [¹⁸F]DCFPyL (piflufolastat) [24,25,26,27] as diagnostic imaging agents have been approved in the US by the Food and Drug Administration (FDA) since 1st December 2020 [28,29,30,31,32] and 27th May 2021 [33,34], respectively. Meanwhile, [¹⁸F]PSMA-1007 [35,36,37,38] is awaiting regulatory approval in European countries pending a phase 3 clinical study being completed (NCT04102553). [^99mTc]Tc-MIP-1404 (trofolostat) is a highly promising SPECT agent [39,40,41,42]. Monographs have been published by the European Directorate for the Quality of Medicines (EDQM) for [⁶⁸Ga]Ga-PSMA-11 (Ph. Eur. Monograph 3044) and [¹⁸F]PSMA-1007 (Ph. Eur. Monograph 3116) which came into force 1st April 2021 and 1st July 2021, respectively [43]. Alberts et al. recently reported an overview that evaluated the most commonly used PET radiopharmaceuticals for the diagnosis of prostate cancer [44].

The most prominent ligands used for PSMA-targeted endoradiotherapy with beta⁻ and alpha emitters are PSMA I&T [45,46,47,48] and PSMA-617 (vipivotide tetraxetan) [49,50,51,52]. The latter radiopharmaceutical is expected to receive marketing authorization in several countries after successfully accomplishing a phase 3 clinical study using [¹⁷⁷Lu]Lu-PSMA-617 (NCT03511664) [53,54,55]. In response to this predicted breakthrough, infrastructural challenges should be addressed in the capacity planning of Nuclear Medicine therapy departments in the future [56]. Moreover, the experience of patients enrolled in a clinical trial involving radioactive treatment [57] and the necessity of a proper patient selection by PET imaging prior endoradiotherapy demands further attention [58,59,60].

In the future, a much broader acceptance and application of PSMA theranostics can be anticipated when diagnosis and treatment are being reimbursed by health insurances [61]. Amongst the aforementioned radiopharmaceuticals, several other radioligands targeting PSMA are either under clinical investigation or have been translated into a clinical setting (

Table 1. PSMA-targeting radiopharmaceuticals marketed, under clinical investigation or in a clinical setting.

Radiopharmaceutical	Brand Name	Radionuclide	Reference	Number of Clinical Trials 1 (Completed/Ongoing)
DCFPyL	PYLARIFY	18F	[24,66] 2	14/51
PSMA-11	Illuccix 3/isoPROtrace-11	68Ga	[20,67] 2,4	24/89
	-	18F	[68,69]	2/-
PSMA-1007	-	18F	[35,70] 4	2/13
PSMA-617	-	44Sc	[71,72]	-/-
		64Cu	[73]	-/-
		68Ga	[49,74]	1/3
		111In	[75]	-/-
		152Tb	[76]	-/-
		177Lu	[49,77]	1/17
		225Ac	[78]	-/2
PSMA I&T	-	68Ga	[45]	-/-
		111In	[79,80]	-/1
		177Lu	[45]	-/5
		225Ac	[48]	-/-
PSMA I&S	-	99mTc	[81]	-/3
MIP-1404	-	99mTc	[39,82]	6/-
rhPSMA-7.3	-	18F	[83,84]	1/3
		177Lu	[85,86]	-/-
JK-PSMA-7	-	18F	[87,88]	-/-
THP-PSMA	GalliProst	68Ga	[89,90]	1/1
PSMA-R2	-	68Ga	-	1/1
		177Lu	-	0/1

¹ According to clinicaltrials.gov accessed on 26 October 2021. ² FDA-approved. ³ TGA-approved. ⁴ Ph. Eur. monograph applicable.

) [19,62]. The combination of radionuclide therapy with other treatment strategies offers new opportunities for patient care, these potential alternatives have been reviewed by Sandhu et al. and Zhang et al. [63,64]. This review aims to provide an overview of novel developments and applications in the area of small-molecule and low-molecular weight-based radiolabeled PSMA inhibitors [65], keeping their prospective clinical translation in mind.

2. Quality and Technical Improvements of PSMA Inhibitors

2.1. Quality Issues and Technical Improvements

Although radiolabeled PSMA inhibitors are used extensively for clinical applications, including radiotracers for which quality standards have been established by legally applicable specific monographs [43], special care needs to be taken during radiosynthesis and quality control of the radiopharmaceuticals.

Iudicello et al. showed that PSMA-11, which is used as a reference standard and precursor for the synthesis of [⁶⁸Ga]Ga-PSMA-11, when dissolved in acidic aqueous solution forms a ^natFe-PSMA-11 complex (a side product with two diastereomeric forms similar to the ones previously reported for [⁶⁸Ga]Ga-PSMA-11) resulting thus, in lower radiochemical yield, chemical purity and radiochemical purity [91,92]. As a consequence, for storage of the PSMA-11 solution, EDTA was added to transchelate the Fe iron from the HBED-CC chelator.

Five-membered ring systems can be formed as radioactive side-products by a thermally mediated and pH-dependent condensation reaction of the Glu-urea-Lys binding motif during synthesis of [¹⁷⁷Lu]Lu-PSMA-617 [93]. These side-products did not show any binding affinity to PSMA and were rapidly excreted via the kidneys from the body of five patients. The synthesis of [¹⁷⁷Lu]Lu-PSMA-617 could be optimized by adjustment of pH and temperature and subsequent reduction of side-products, however, the risk of formation of unwanted radioactive side-products for PSMA inhibitors bearing the Glu-urea-Lys binding motif still remains [93].

Reaction temperature during radiolabeling plays an important role as shown above. Attempts have been made to replace DOTA, which requires time and heating for complexing radiometals, with a different chelator that allows low temperature radiolabeling in particular for the alpha-particle emitting isotope ²²⁵Ac. Thiele et al. showed that the eighteen-membered macrocyclic chelator macropa is suitable to complex ²²⁵Ac at room temperature (RT) in 5 min with high stability in human serum over 8 days [94]. Macropa was then conjugated to the PSMA/albumin binding ligand RPS-070 and radiolabeled with ²²⁵Ac within 20 min at RT. [²²⁵Ac]Ac-macropa-RPS-070 was tested in LNCaP tumor bearing mice where it showed high in vivo stability and high kidney uptake after 4 h (52 ± 16% ID/g), which then cleared rapidly, while tumor uptake (13 ± 3% ID/g) washout was slower. Reissig et al. expanded the “macropa”-principle by adding one or two alkyne moieties for conjugation of biomolecules using the copper-catalyzed azide-alkyne-cycloaddition [95]. A PSMA binding block was attached via click chemistry to obtain either the monomer mcp-M-PSMA or the dimer mcp-D-PSMA. Radiolabeling was done within 1 h at RT, both [²²⁵Ac]Ac-mcp-M-PSMA and [²²⁵Ac]Ac-mcp-D-PSMA were evaluated in vitro and in vivo using LNCaP cells and the corresponding xenograft mouse model. It is noticeable that the tumor uptake for the monomer was higher within the first hour and then decreased while the tumor uptake for the dimer constantly increased within 24 h making it more suitable for a therapy approach [95].

Radiopharmaceuticals are also prepared via so-called cold kits, for a long time established in Nuclear Medicine by reconstitution with the eluate of ^99mTc generators [96]. More recently, this approach has seen a wider application also for PET radiopharmaceuticals with three commercially available kits using either PSMA-11 (Illuccix and isoPROtrace-11) or THP-PSMA (GalliProst) for radiolabeling with the eluate of a ⁶⁸Ge/⁶⁸Ga generator [97]. Illuccix has been approved by the Australian Therapeutic Goods Administration (TGA) on 2nd November 2021 [98].

2.2. Salivary Gland Uptake

Uptake in the salivary glands is the dose-limiting factor during endoradiotherapy with small-molecule PSMA inhibitors, resulting in (partially reversible) xerostomia when applied with ¹⁷⁷Lu, and severe to persistent xerostomia when applied with ²²⁵Ac [50,51]. The quality of life for patients can, therefore, be significantly impaired despite preventive strategies, e.g., injection of botulinum toxin, external cooling, or gustatory stimulation have been described [99,100,101,102,103]. The uptake mechanism of PSMA inhibitors in salivary glands has not been thoroughly investigated to date, a more comprehensive review concerning known mechanisms is described by Heynickx et al. [103]. Tönnesmann et al. showed in quantitative autoradiography experiments that in the salivary glands of pigs for small-molecule and low-molecular weight PSMA inhibitors the uptake is highly non-specific and at the same time specific while the exact ratio is not yet understood [104]. Attempts to minimize the uptake of the PSMA-targeted radiopharmaceutical in the salivary glands have been made following the modifications at the inhibitor component. This resulted in low salivary gland uptake and very low tumor uptake [105]. Two independent preclinical studies demonstrated that additional substance amount-dependent administration of nonradioactive PSMA ligand (PSMA-11 or DCFPyL, respectively) and reduction of the effective molar activity significantly reduces the radioligand uptake ([¹⁷⁷Lu]Lu-PSMA-617 or [¹⁸F]DCFPyL, respectively) in the salivary glands (

Table 2. Biodistribution data of [¹⁷⁷Lu]Lu-PSMA-617 (5 pmoles) with co-injected non-radioactive PSMA-11 for tumor and selected non-target organs one hour after injection in athymic mice bearing PC-3 PIP tumors [106].

Added Amount of “Cold” PSMA-11 (Pmoles)	Tumor Uptake (% ID/g)	Kidney Uptake (% ID/g)	Salivary Gland Uptake (% ID/g)
0	21.71 ± 6.13	123.14 ± 52.52	0.48 ± 0.11
5	18.70 ± 2.03	132.31 ± 47.40	0.45 ± 0.15
100	26.44 ± 2.94	84.29 ± 78.25	0.38 ± 0.30
500	16.21 ± 3.50	2.12 ± 1.88	0.08 ± 0.03
1000	13.52 ± 3.68	1.16 ± 0.36	0.09 ± 0.07
2000	12.03 ± 1.96	0.64 ± 0.23	0.05 ± 0.02

) [106,107]. However, these promising results need to be confirmed in further preclinical and clinical studies.

3. PSMA Inhibitors for Guided Surgery

3.1. PSMA Inhibitors for Radioguided Surgery

In addition to the properties for noninvasive single photon emission computed tomography (SPECT) or positron emission tomography (PET) imaging, the radiation exposed from the radionuclide bound to the PSMA inhibitor can also be used for the detection of tumor entities during intraoperative surgery using a gamma probe in a practice named radioguided surgery (RGS) [108].

PSMA I&T and PSMA-617 radiolabeled with ¹¹¹In were initially used for SPECT imaging and RGS [75,79,80,109,110], with both showing detection rates lower than [⁶⁸Ga]Ga-PSMA-11 [111]. However, due to the limited availability of [¹¹¹In]InCl₃, PSMA I&T derived Mas₃-y-nal-k(Sub-KuE) (PSMA I&S) was developed and radiolabeled with ^99mTc [81]. [^99mTc]Tc-PSMA I&S RGS using a gamma probe was applied in a clinical setting after identifying tumor lesions with [⁶⁸Ga]Ga-PSMA-11 PET [112,113]. In a study including 210 PCa patients, high detection rates were found for [^99mTc]Tc-PSMA I&S for PSA levels > 4 ng/mL [114]. The radiation dosimetry of [^99mTc]Tc-PSMA I&S, which was performed in a single-center study, resulted in comparable effective doses with other ^99mTc radiopharmaceuticals and lower effective doses than either ⁶⁸Ga- or ¹⁸F-radiolabeled PSMA inhibitors ([^99mTc]Tc-PSMA I&S: 0.0052 mSv/MBq, [⁶⁸Ga]Ga-PSMA-11: 0.0236 mSv/MBq, [¹⁸F]PSMA-1007: 0.022 mSv/MBq) [115,116,117].

As an undesirable consequence of RGS, personnel are typically exposed to constant radiation from the radiopharmaceutical. The occupational radiation exposure for workers involved in RGS using [^99mTc]Tc-PSMA I&S was determined with a maximum dose of 0.32 mSv/year for 100 procedures [118]. When laparoscopic gamma probes are used for sentinel lymph node dissection [119] they suffer from practical limitations, due to rotational impairment of the probe [120]. This limitation has been circumvented by the development of a drop-in probe, which can be controlled by grasps of the DaVinci robotic-assisted surgery system, thereby providing additional rotational freedom and reducing the radiation dose [121]. A first-in-human application has been reported for [^99mTc]Tc-PSMA I&S after preoperative [⁶⁸Ga]Ga-PSMA-11 PET/CT [122]. In an experimental study with seven patients, the drop-in concept was tested with [⁶⁸Ga]Ga-PSMA-11 using a beta probe instead of a gamma probe. After an additional dose of approximately 70 MBq [⁶⁸Ga]Ga-PSMA-11 per patient prior surgery, the obtained surgical specimens were evaluated ex vivo using a DaVinci robotic system equipped with the drop-in beta probe [123]. Radiation dose to workers limits the use to 62 procedures/year of this technique, but improvements towards a lower occupational radiation dose are feasible. Alternative imaging technologies focusing on beta-radiation emission, e.g., Cerenkov luminescence imaging using [⁶⁸Ga]Ga-PSMA-11 are currently under early clinical investigation [124,125,126,127,128,129,130].

3.2. PSMA Inhibitors for Fluorescence-Guided Surgery and/or Targeted Photodynamic Therapy

Aside from radionuclide detection for radioguided surgery, an attractive alternative for intraoperative guided surgery is radiolabeled molecules functionalized with fluorescent dyes for optical/fluorescence imaging [131,132,133,134]. A variety of commercially available dyes with different absorptions and emission spectra used for dual-labeled imaging agents has been reported [135]. Dual-labeled agents allow preoperative nuclear imaging (prestaging/staging) and subsequent treatment planning while fluorescence imaging makes surgical tumor resection possible and more accurate due to the high spatial resolution.

Baranski et al. studied the PSMA-11 derived hybrid compound Glu-urea-Lys-HBED-CC-IRDye800CW in vitro and in vivo in LNCaP-tumor bearing mice and healthy pigs [136]. [⁶⁸Ga]Ga-Glu-urea-Lys-HBED-CC-IRDye800CW showed an increased tumor uptake in mice compared to [⁶⁸Ga]Ga-PSMA-11 (13.66 ± 3.73% ID/g vs. 4.89 ± 1.34% ID/g) [136,137]. Non-radiolabeled Glu-urea-Lys-HBED-CC-IRDye800CW has been investigated in healthy pigs using a DaVinci robotic system. A PSMA-specific fluorescence signal has been shown in vivo and ex vivo for both mice and pigs. More recently, the linker design of Glu-urea-Lys-HBED-CC-IRDye800CW was modified by insertion of histidine- and glutamic acid spacers ((HE)₃) near the PSMA binding motif to further reduce background uptake in non-target tissue and to accelerate excretion [138,139]. Glu-urea-Lys-(HE)₃-HBED-CC-IRDye800CW (PSMA-914) was used in a proof-of-concept study, with a first, pre-operative PET/CT imaging, 1-day prior surgery in a patient with high-risk prostate carcinoma and a second PET/CT imaging, after the surgical removal of the primary tumor [140]. ⁶⁸Ga-radiolabeled PSMA-914 and PSMA-914 alone were administered 1-h prior to PET imaging and DaVinci robot-assisted prostatectomy respectively.

Schottelius et al. conjugated the disulfonated analog of the far-red cyanine dye Cy5 to PSMA I&T to obtain the hybrid imaging agent PSMA I&F (DOTAGA-D-Lys(N_ε-Sulfo-Cy5)-D-(3-iodo)Tyr-D-(3-iodo)Tyr-D-Lys(N_ε-Sub-KuE) [141]. Tumor uptake in LNCaP xenograft-bearing mice was comparable between [⁶⁸Ga]Ga-PSMA I&F and [⁶⁸Ga]Ga-PSMA I&T (4.5 ± 1.8% ID/g vs. 4.9 ± 1.6% ID/g) while kidney uptake was almost doubled for [⁶⁸Ga]Ga-PSMA I&F (105.8 ± 22.7% ID/g vs. 53.3 ± 9.0% ID/g).

One limitation/challenge regarding the utilization of fluorescent dyes is that the detection efficacy of tumor lesions is limited to superficial tissue [142,143]. Certain dyes can also be used for photodynamic therapy [144]. In this approach, a dye can also simultaneously act as a photosensitizer which is selectively activated with long wavelength light, subsequently causing reactive oxygen species (ROS) formation leading to the induction of immediate cell death [108]. Derks et al. synthesized a variety of Glu-urea-Lys-based PSMA-targeting ligands conjugated to IRDye700DX [145]. The most promising candidate PSMA-N064 consists of a DOTAGA chelator, additional glutamic acid residues in the linker structure, a Glu-urea-Lys binding motif, and a fluorophore. After radiolabeling of PSMA-N064 with ¹¹¹In, an increased tumor uptake of up to 2 h post injection (13.1 ± 2.3% ID/g) was determined in mice bearing LS1754T colon carcinoma cells transfected with human PSMA (LS1754T-PSMA). The tumors could be visualized using SPECT/CT and near infrared fluorescence scanners. Interestingly, the incorporation of the fluorophore leads to a higher internalization in tumor cells than the corresponding ligand without the dye (15.1 ± 0.8% ID/g vs. 6.7 ± 1.1% ID/g). Accordingly, the fluorophore parts may have a positive impact on tumor uptake [136,145]. Samples obtained from biopsies of human normal tissue could be distinguished from tumor tissue after incubation with PSMA-N064 using fluorescence imaging. In a proof-of-concept study for targeted photodynamic therapy, the incubation of LS1754T-PSMA cells with PSMA-N064 and exposure to NIR light resulted in reduced cell viability (34 ± 3.2%).

4. Multitargeted PSMA Inhibitors

4.1. PSMA + Albumin

PSMA inhibitors used for endoradiotherapy, such as PSMA-617 and PSMA I&T are rapidly excreted via the kidneys, increasing the radiation exposure to these organs [146]. Introducing albumin-binding groups into the chemical structure of small molecules can increase their blood circulation time, which can potentially lead to enhanced tumor uptake, while the reduction of injected activity results in lower non-target tissue doses [147,148].

The 4-(p-iodophenyl)acetic acid moiety was employed as an albumin-binding group by Kelly et al. in radiotracers bearing a DOTA chelator and a urea-based PSMA binding-motif radiolabeled with ¹³¹I (RPS-025) and ¹⁷⁷Lu (RPS-063 & RPS-067). This modification lead to a 4-fold higher tumor uptake compared to [¹⁷⁷Lu]Lu-PSMA-617, however, kidney-uptake was still significantly high [149,150]. It was observed that the use of polyethylene glycol (PEG) linkers of varying lengths had a dramatic impact on PSMA binding affinity as well as blood clearance [150]. Replacement of the albumin-binding group 4-(p-iodophenyl)acetic acid moiety with 4-(p-iodophenyl)butanoic acid (RPS-072) further reduced radiation exposure to the kidney and blood clearance [151].

Benešová et al. used the 4-(p-iodophenyl)butanoic acid moiety as an albumin-binding group in a similar way and extended the linker entity with aspartate residues to counterbalance the lipophilicity of the albumin-binding group [152]. Three radioligands consisting of a DOTA chelator, the aforementioned linker, the albumin-binding group and a urea-based PSMA binding motif (PSMA-ALB-02, PSMA-ALB-05 and PSMA-ALB-97) were radiolabeled with ¹⁷⁷Lu and evaluated against [¹⁷⁷Lu]Lu-PSMA-617 in PC-3 PIP tumor bearing mice with all three ligands showing a high tumor-to-blood ratio but still high kidney uptake. These results were further optimized by the replacement of the 4-(p-iodophenyl)butanoic acid moiety with (p-tolyl)butanoic acid [153], as the latter demonstrated a reduced albumin-binding affinity [147]. The most promising candidate PSMA-ALB-56 was compared against PSMA-617 in a therapy study of PC-3 PIP tumor bearing mice. [¹⁷⁷Lu]Lu-PSMA-ALB-56 showed a better median survival time of the treated mice than [¹⁷⁷Lu]Lu-PSMA-617 when the same activity was applied (5 MBq/mouse), with 4/6 of the [¹⁷⁷Lu]Lu-PSMA-ALB-56-treated mice surviving the study. The use of 2-PMPA improved the tumor-to-kidney ratio by reducing kidney uptake of [¹⁷⁷Lu]Lu-PSMA-ALB-56 [154]. In a recently published clinical study, ten patients with metastatic castration resistant prostate cancer (mCRPC) received [¹⁷⁷Lu]Lu-PSMA-ALB-56 endoradiotherapy with a higher absorbed dose in tumor lesions and similar salivary gland-uptake compared to PSMA-617 and PSMA I&T [155]. Nevertheless, kidney and red marrow uptake in these patients remains high, this demands further preclinical optimization of albumin-binding PSMA radioligands (e.g., upon utilization of ibuprofen as an albumin-binding entity) [156,157,158].

4.2. PSMA + GRPr

Apart from PSMA, gastrin-releasing peptide receptors (GRPr) are frequently highly expressed in PCa tumors. The bombesin peptide (BBN = Pyr¹-Gln²-Arg³-Leu⁴-Gly⁵-Asn⁶-Gln⁷-Trp⁸-Ala⁹-Val¹⁰-Gly¹¹-His¹²-Leu¹³-Met¹⁴-NH₂) is considered a potent agonist for GRPr, while over the years several BBN analogs (either agonists or antagonists of GRPr) have been radiolabeled and evaluated for potential application in Nuclear Medicine [159,160]. The expression of both receptors in tumors is heterogeneous [161], a fact that could potentially result in reduced image quality or false-negative results. A recent preclinical study, which evaluated retrospectively frozen prostatectomy samples (n = 20) with tissular microimaging using [¹¹¹In]In-PSMA-617 and the GRPr antagonist [¹¹¹In]In-RM2 ([¹¹¹In]In-DOTA-4-amino-1-carboxymethyl-piperidine-D-Phe⁶-Gln⁷-Trp⁸-Ala⁹-Val¹⁰-Gly¹¹-His¹²-Sta¹³-Leu¹⁴-NH₂), showed that for [¹¹¹In]In-PSMA-617 the binding was high and independent of the metastatic risk (p = 0.665), Gleason score (p = 0.555), or PSA value (p = 0.404), while [¹¹¹In]In-RM2 exhibited a significantly higher binding in the low metastatic risk group (p = 0.046), in the low PSA value group (p = 0.001), and in samples with Gleason 6 score (p = 0.006) [162]. A limited number of patients have been screened with a PSMA-targeting imaging agent ([⁶⁸Ga]Ga-PSMA-11 or [¹⁸F]DCFPyL) and the GRPr antagonist [⁶⁸Ga]Ga-RM2 [163,164,165,166,167] resulting rather in a complementing than in a competing approach of PSMA vs. GRPr imaging [165,168].

However, in order to overcome this limitation, and increase specificity and sensitivity of the imaging method, the heterodimers approach has been described for a combination of various targets [169,170]. Specifically, heterodimers (Hd) targeting PSMA/GRPr have been recently reviewed by Liolios et al. [171]. The first reported heterodimer targeting both PSMA and GRPr consisted of the acyclic chelator HBED-CC, which was linked to either side via its two carboxylic groups not participating at the complexation of the radiometal with the Lys-CO-Glu-OH PSMA binding motif and the GRPr agonist H₂N-PEG₂-[D-Tyr⁶,beta-Ala¹¹,Thi¹³,Nle¹⁴)BN(6–14) [172]. The BBN analog, in this case, was structurally relevant to [⁶⁸Ga]GaBZH₃ (DOTA-PEG₂-[D-tyr⁶, beta-Ala¹¹, Thi¹³, Nle¹⁴] BN(6–14)), which contained a modified BBN sequence, resistant to peptidases, and which has been clinically tested [173,174]. The heterodimer Glu-urea-Lys(Ahx)-HBED-CC-BZH3 (Hd-1) showed affinity values comparable to the monomers (PSMA IC₅₀ = 25 ± 5.4 nM, GRPr IC₅₀ = 9.0 ± 1.8 nM) during in vitro testing in PC-3 (GRPr-positive, androgen independent) and LNCaP (PSMA-positive, androgen depended) cells. Specific cell-binding (either surface bound or internalized) for the ⁶⁸Ga labeled heterodimer was similar to the control ⁶⁸Ga-monomers, while tumor uptake in LNCaP and AR42J (GRPr positive) xenografts were 5.44 ± 1.54% ID/g and 3.34 ± 1.05% ID/g, respectively at 1 h p.i. Tumors were clearly visible during micro-PET imaging (1 h p.i.), however, a high kidney uptake was noted (110.46 ± 8.80% ID/g, 1 h, p.i.), which was minimized during PSMA block (16.04 ± 3.94% ID/g), but remained unaffected during GRPr block (109.52 ± 34.52% ID/g) clearly indicating the role of PSMA receptors regarding kidney accumulation. This heterodimer was improved to reduce kidney uptake by the addition of so-called (HE)_n (n = 1–3) amino acid spacers between the HBED-CC chelator and the PSMA pharmacophore [138]. The new heterodimers (HE_n, where n = 0, Hd-2, n = 1, Hd-3, n = 2, Hd-4, n = 3, Hd-5), when tested in vitro they showed high affinity in all cell lines tested: PC-3 (IC₅₀ = 4.40 to 7.72 nM), AR42J (IC₅₀ = 2.58 to 5.06 nM) and LNCaP (IC₅₀ = 17.4 to 42.4 nM), which were comparable to the respective monomers for each case (PC-3, 3.65 nM, AR42J, 1.29 nM, LNCaP 7.5 nM). During the in vivo experiments in mice with PC-3 ad LNCaP xenografts the HE₂ spacer in addition to the reduction of kidney uptake, also managed to increase tumor uptake for both xenografts (Table 3).

Bandari et al. synthesized and radiolabeled with the PET radionuclide ⁶⁴Cu, the heterodimer: DUPA-6-Ahx-(⁶⁴Cu-NODAGA)-5-Ava-BBN(7–14)NH₂ [175]. The heterodimer (Hd-6) was evaluated in vitro in PC-3 and LNCaP cells, where it showed high affinity (PC-3: IC₅₀ = 11.1 ± 0.46 nM; LNCaP: IC₅₀ = 1.16 ± 1.35 nM), while during in vivo MicroPET scintigraphy the xenografted PC-3 and LNCaP tumors were clearly visible at 18 h p.i. Bandari et al. also synthesized the heterodimer [DUPA-6-Ahx-Lys(DOTA)-6-Ahx-RM2] using the antagonist BN peptide (BBN ANT) sequence of RM2 (BBN ANT = D-Phe⁶-Gln⁷-Trp⁸-Ala⁹-Val¹⁰-Gly¹¹-His¹²-Sta¹³-Leu¹⁴-NH₂) (Hd-7), which was labeled with ⁶⁸Ga, ¹¹¹In, or ¹⁷⁷Lu for theranostic applications [176]. In a later study and based on this structure, Bandari et al. synthesized and evaluated several heterodimeric ligands testing different spacer groups in addition to 6-Ahx connecting the BBN peptide with the chelator DO3A (DUPA-6-Ahx-[DO3A]-X-BBN ANT, where DO3A = 1,4,7,10-tetraazacyclododecane-1,4,7-triacetic acid and X = 5-aminovaleric acid, 5-Ava; 6- aminohexanoic acid, 6-Ahx; 8-aminooctanoic acid, 8-Aoc; and para-aminobenzoic acid, AMBA). In vitro testing revealed low IC₅₀ values for PC-3 (4–7 nM), and LNCaP cells (10–16 nM), both before and after metalation with ^natIn and ^natLu [177]. Among the different candidates, the heterodimer DUPA-6-Ahx-[DO3A]-8-Aoc -BBN ANT (Hd-8) was selected as the optimal candidate for further in vivo testing with PC-3 and LNCaP xenografts. The selection criterion was its higher accumulation in the pancreas (5.45 ± 0.47% ID/g, 1 h p.i.) during biodistribution studies in normal mice, which clearly indicated GRPr targeting since the normal pancreas is known to overexpress GRPr receptors. The selected heterodimer after radiolabeling with ¹¹¹In and ¹⁷⁷Lu presented high stability, both in PBS and in human serum after 4 h of incubation (at RT), while minor radiolytic degradation was observed at the 24 h. In vivo biodistribution experiments in mice carrying PC-3 or PC-3 PIP xenografts showed rapid elimination (85% of the injected dose, 4 h p.i.) of each tracer (¹¹¹In or ¹⁷⁷Lu) from the bloodstream with excretion profiles being primarily via the renal-urinary pathway. Kidney uptake was also high ranging from 8.90 ± 1.40% ID/g to 32.2 ± 28.8% ID/g (until 4 h p.i.), which was also expected considering PSMA expression in this organ. Tracer uptake in xenografted tumors ranged from 4.74 ± 0.90 to 7.51 ± 2.61% ID/g, with the highest values observed at 1 h p.i. and gradually decaying thereafter. Higher tumor values were observed for the ¹⁷⁷Lu labeled compound in both xenografts, which according to the authors was due to the presence of more nonradioactive [DUPA-6-Ahx-([^natLu]-Lu-DO3A)-8-Aoc-BBN ANT] in comparison to the ¹¹¹In labeled analog [177]. Possibly, the nonradioactive ligand blocked the clearance route of the radioactive tracer, resulting in higher kidney radioactivity and more importantly in higher circulating amounts of [DUPA-6-Ahx-([¹⁷⁷Lu]-Lu-DO3A)-8-Aoc-BBN ANT], which eventually was accumulated in the tumors. This hypothesis in combination with the increased radioactivity amounts observed in normal organs during the blocking experiments further noted the importance of high specific activity for such radiopharmaceutical applications and further enhancement of heterodimeric ligands, i.e., with albumin binders.

Mendoza-Figueroa et al. designed a heterodimeric ⁶⁸Ga-labeled radiotracer ([⁶⁸Ga]Ga-iPSMA-BN) (Hd-9) consisting of an iPSMA (Nal-Lys-CO-Glu-OH) and a Lys³-Bombesin part linked to a DOTA chelator. When [⁶⁸Ga]Ga-iPSMA-BN was evaluated against the monomers [⁶⁸Ga]Ga-iPSMA and [⁶⁸Ga]Ga-BN in vitro and in vivo in LNCaP cells and PC-3 cells, it showed superiority against each monomer [⁶⁸Ga]Ga-iPSMA and [⁶⁸Ga]Ga-BN, in both cell lines and animal models (Table 3) [178]. The same group has published a study where they evaluated this heterodimer labeled with ¹⁷⁷Lu [179]. The heterodimer showed a significant decrease in LNCaP (10.15%) and PC-3 (40.10% at 48 h) viability in vitro, and a high LNCaP and PC-3 tumor uptake in ex vivo biodistributions (5.21 and 3.21% ID/g at 96 h, respectively), and Micro-SPECT/CT imaging studies (SUV_max of 1.93 ± 0.30 and 1.76 ± 0.10 in LNCaP and PC-3, respectively), possibly influenced by the heterobivalent effect. More recently, biokinetics and dosimetry data have been obtained for [⁶⁸Ga]Ga-iPSMA-BN in a study of four healthy patients showing specific uptake in the pancreas (GRPr expressing) and salivary glands (PSMA expressing) [180].

Mitran et al. synthesized the heterodimer NOTA-DUPA-RM26 (BQ7800) consisting of (Hd-10) (DUPA = 2-[3-(1,3-dicarboxypropyl)-ureido]pentanedioic acid) consisting of a Glu-CO-Glu PSMA-binding motif and the GRPr antagonist RM26 (D-Phe-Gln-Trp-Ala-Val-Gly-His-Sta-Leu-NH2). Both binding motifs were coupled via glutamic acid and contain NOTA as a chelator for radiolabeling with either ⁶⁸Ga or ¹¹¹In [181]. In vitro preclinical evaluation resulted in high affinity for GRPr (IC₅₀ = 4.0 ± 1.0 nM) and rather low for PSMA (IC₅₀ = 824 ± 230 nM). However, at 1 h p.i., tumor uptake in PC-3 PIP xenografts for [¹¹¹In]In-BQ7800 (12 ± 2% ID/g) and [⁶⁸Ga]Ga-BQ7800 (8 ± 2% ID/g) was higher than the one in normal organs and was decreased thereafter. However, kidney uptake was similar to the tumor uptake, while increased uptake was observed in organs expressing GRPr receptors. Attempts to improve the PSMA-specific tumor uptake were made by incorporating phenylalanine into the linker of the PSMA-binding motif and shortening of the PEG linker (BQ7812). The tumor uptake of [¹¹¹In]In-BQ7812 (Hd-11) was doubled compared to [¹¹¹In]In-BQ7800, but the modifications also led to a three-fold increase of uptake in the kidneys [182].

Finally, three additional bispecific heterodimers based on RM2 peptide sequence (D-Phe⁶-Gln⁷-Trp⁸-Ala⁹-Val¹⁰-Gly¹¹-His¹²-Sta¹³-Leu¹⁴-NH₂) and PSMA-617 are reported [183]. The two pharmacophores were linked via the spacer: X-triazolyl-Tyr-PEG₂, where X = 0 (Hd-12), PEG₂ (Hd-13), (CH₂)₈ (Hd-14). The resulting heterodimers were radio-iodinated and evaluated in vitro for binding specificity, cellular retention, and affinity. In vivo specificity and tumor uptake for all heterodimers was studied in PC-3 and LNCaP xenografts (Table 3), while [¹²⁵I]I-Hd-13 was also evaluated in PC-3 PIP xenografts where it showed high tumor accumulation (30– 35% ID/g at 3 h p.i.). However, this was followed by high kidney radioactivity values like the first heterodimer: [⁶⁸Ga]Ga-Hd-1.

Table 3. Heterodimers targeting PSMA/GRPr, tumor uptake (1 h p.i.) expressed as (% ID/g) in PC-3, AR42J, LNCaP or PC-3 PIP xenografts (Mean value ± Std).

Heterodimer	Tumor Uptake (1 h p.i.; &24 h p.i., # 96 h p.i.) (% ID/g ± SD)				Ref.
	PC-3	AR42J	LNCaP	PC-3 PIP
[68Ga]Ga-Hd-1	-	3.34 ± 1.04	5.44 ± 1.54	-	[172]
[68Ga]Ga-Hd-2	0.84 ± 0.18	-	2.38 ± 0.05	-	[138]
[68Ga]Ga-Hd-3	2.67 ± 1.25	-	2.41 ± 1.24	-
[68Ga]Ga-Hd-4	2.23 ± 0.27	-	10.66 ± 4.19	-
[68Ga]Ga-Hd-5	1.66 ± 0.41	-	3.22 ± 0.22	-
[64Cu]Cu-Hd-6	-	-	-	-	[175]
Hd-7	-	-	-	-	[176]
([111In]In-Hd-8	4.74 ± 0.90	-	-	5.38 ± 1.07	[177]
[177Lu]Lu-Hd-8	7.51 ± 2.61	-	-	7.37 ± 2.89	[177]
[68Ga]Ga-Hd-9	7.39 ± 0.65	-	13.67 ± 3.88	-	[178]
[177Lu]Lu-Hd-9	1.76 ± 0.10 & ~ 3.21 #	-	1.93 ± 0.3 & ~ 5.21 #	-	[179]
[111In]In-Hd-10	-	-	-	12.0 ± 2.0	[181]
[68Ga]Ga-Hd-10	-	-	-	8.0 ± 2.0	[181]
[111In]In-Hd-11	-	-	-	16.10 ± 2.96	[182]
[125I]I-Hd-12	3.0 ± 0.3		11.0 ± 3.0		[183]
[125I]I-Hd-13	4.3 ± 0.6		10.0 ± 3.0
[125I]I-Hd-14	2.2 ± 0.6		3.0 ± 2.0	~ 21.0

Table 3. Heterodimers targeting PSMA/GRPr, tumor uptake (1 h p.i.) expressed as (% ID/g) in PC-3, AR42J, LNCaP or PC-3 PIP xenografts (Mean value ± Std).

Heterodimer	Tumor Uptake (1 h p.i.; &24 h p.i., # 96 h p.i.) (% ID/g ± SD)				Ref.
	PC-3	AR42J	LNCaP	PC-3 PIP
[68Ga]Ga-Hd-1	-	3.34 ± 1.04	5.44 ± 1.54	-	[172]
[68Ga]Ga-Hd-2	0.84 ± 0.18	-	2.38 ± 0.05	-	[138]
[68Ga]Ga-Hd-3	2.67 ± 1.25	-	2.41 ± 1.24	-
[68Ga]Ga-Hd-4	2.23 ± 0.27	-	10.66 ± 4.19	-
[68Ga]Ga-Hd-5	1.66 ± 0.41	-	3.22 ± 0.22	-
[64Cu]Cu-Hd-6	-	-	-	-	[175]
Hd-7	-	-	-	-	[176]
([111In]In-Hd-8	4.74 ± 0.90	-	-	5.38 ± 1.07	[177]
[177Lu]Lu-Hd-8	7.51 ± 2.61	-	-	7.37 ± 2.89	[177]
[68Ga]Ga-Hd-9	7.39 ± 0.65	-	13.67 ± 3.88	-	[178]
[177Lu]Lu-Hd-9	1.76 ± 0.10 & ~ 3.21 #	-	1.93 ± 0.3 & ~ 5.21 #	-	[179]
[111In]In-Hd-10	-	-	-	12.0 ± 2.0	[181]
[68Ga]Ga-Hd-10	-	-	-	8.0 ± 2.0	[181]
[111In]In-Hd-11	-	-	-	16.10 ± 2.96	[182]
[125I]I-Hd-12	3.0 ± 0.3		11.0 ± 3.0		[183]
[125I]I-Hd-13	4.3 ± 0.6		10.0 ± 3.0
[125I]I-Hd-14	2.2 ± 0.6		3.0 ± 2.0	~ 21.0

4.3. PSMA + Integrin α_vβ₃

Two heterobivalent imaging agents targeting PSMA and integrin-α_vβ₃ surface markers, both overexpressed in certain tumor epithelium and/or neovasculature were synthesized and evaluated by Shallal et al. [184]. Integrins α_vβ₃ are considered an ideal target for the development of radioligands, since they are overexpressed on tumor vasculature due to angiogenesis, and on the cell membrane in various tumors, i.e., ovarian cancer, neuroblastoma, breast cancer and melanoma [185]. The tripeptide RGD (arginine-glycine-aspartic) amino acid sequence and its cyclic RGD (cRGD) analogs specifically bind to the integrin α_vβ₃ receptor and thus have provided the platform for the development of various radioligands [185]. The heterodimers consisted of Lys-CO-Glu-OH PSMA binding motif connected to the Lys of a cyclic cRGDfK peptide (cyclo-(Arg-Gly-Asp-D-Phe-Lys)) and coupled either to the DOTA chelator for radiolabeling (Hd-15) or to the dye IRDye800CW (Hd-16). In vitro testing in PC-3 PIP and U87-MG cells and in isolated proteins showed low affinity with high IC₅₀ values for the both PSMA and α_vβ₃ receptors, i.e., Hd-15: PC-3 PIP, IC₅₀ = 479 nM, U87-MG, IC₅₀ = 1536 nM, Hd-16: integrin α_vβ₃ IC₅₀ = 90 nM. In vivo, optical imaging and ex vivo biodistribution studies for Hd-16 showed specific tumor uptake for PC-3 PIP tumors expressing PSMA, and U87-MG tumors expressing integrin α_vβ₃. Tumor uptake was dose-dependent, and even at the lowest dose (0.5 nmol) the tumors were clearly visible. However, this study did not include a metalated or a radiolabeled version of Hd-15 [184].

5. Future Methods and Concepts for PSMA Inhibitor Radiolabeling

The fluoroglycosylation method by Maschauer et al. [186] has been applied to obtain 2-[¹⁸F]FGlc-PSMA and 6-[¹⁸F]FGlc-PSMA with the latter example highlighting a 10-fold decrease in kidney uptake in PC-3 PIP tumor bearing mice compared to 2-[¹⁸F]FGlc-PSMA and [⁶⁸Ga]Ga-PSMA-11 [187].

Greifenstein et al. used a squaric acid (SA) moiety to link the Glu-urea-Lys PSMA binding motif with the chelator [188]. Out of three SA-PSMA inhibitors with different chelators, DOTAGA.SA.PSMA emerged as a lead structure to be compared against ⁶⁸Ga-radiolabeled PSMA-617 and PSMA-11. [⁶⁸Ga]Ga-DOTAGA.SA.PSMA showed tumor uptake in LNCaP tumor bearing mice comparable with [⁶⁸Ga]Ga-PSMA-617 and lower than [⁶⁸Ga]Ga-PSMA-11 (5.6 ± 0.3% ID/g vs. 6.5 ± 1.0% ID/g vs. 12.8 ± 1.5% ID/g), but significantly lower uptake in kidneys (3.2 ± 0.5% ID/g vs. 16.2 ± 5.7% ID/g vs. 210.8 ± 8.1% ID/g) and in other organs (60 min p.i.) showing thus a potential as therapeutic agent [189].

The isotopic exchange chemistry for the fluorine isotopes ¹⁹F/¹⁸F has been described by Schirrmacher et al. using silicon-fluoride-acceptor (SiFA) building blocks [190]. This concept has been adopted for PSMA inhibitors, which also introduce a chelator into the molecule in close proximity to the SiFA motif in order to increase hydrophilicity [191]. This permits the radiolabeling of the molecule with either ¹⁸F at the SiFA building block or with a radiometal of choice (e.g., ⁶⁸Ga, ¹⁷⁷Lu, ²²⁵Ac) at the chelator while the moiety not used for radiolabeling is occupied with the non-radioactive ¹⁹F or metal, which has been called by the authors radiohybrid PSMA (rhPSMA) inhibitors. Wurzer et al. developed a series of these rhPSMA inhibitors, with rh-PSMA-7 being the most promising candidate being radiolabeled with ¹⁸F, bearing ^natGa within the chelator and being clinically evaluated in a number of patients with primary and biochemical recurrence of PCa [191,192,193,194]. Further improvement of rh-PSMA-7 has been achieved by investigation of its four stereoisomers and their influence on pharmacokinetics [83]. The resulting lead candidate [¹⁸F]Ga-rh-PSMA-7.3 showed the highest tumor accumulation and low uptake in non-target tissues. This was confirmed in healthy volunteers determining biodistribution and radiation dosimetry data (NCT03995888) and in a larger cohort of patients [84,195,196,197]. The molar activity of [¹⁸F]Ga-rh-PSMA-7.3 had a minor influence on the biodistribution [198,199]. Three clinical trials using [¹⁸F]Ga-rh-PSMA-7.3 are either ongoing or will commence in the near future (NCT04978675/NCT04186819/NCT04186845). The therapeutic counterpart [¹⁷⁷Lu]Lu-rhPSMA-7.3 showed a higher anti-tumor effect and longer median survival in C4-2 xenograft-bearings SCIDS than [¹⁷⁷Lu]Lu-PSMA I&T [85] followed by a pre-therapeutic dosimetry study in six mCRPC patients confirming the preclinical results [86].

Another approach of isotopic exchange has been reported for eight PSMA inhibitors by substituting ¹⁹F with ¹⁸F at a trifluoroborate group [200]. This concept has been extended to the radiohybrid ligand DOTA-AMBF₃-PSMA [201]. The PSMA-617 pharmacophore was used as a starting point, then a lysine-trifluoroborate moiety was introduced and linked to a DOTA-chelator allowing radiolabeling with either ¹⁸F or a radiometal. DOTA-AMBF₃-PSMA radiolabeled with ¹⁸F/free, ¹⁸F/^natCu, ^natF/⁶⁴Cu and ^natF/¹⁷⁷Lu was preclinically evaluated in LNCaP xenografts with tumor uptake > 10% ID/g for all radiotracers. [⁶⁴Cu]Cu-DOTA-AMBF₃-PSMA showed increased liver uptake, most likely due to transchelation of ⁶⁴Cu from the DOTA-chelator which has also been observed for [⁶⁴Cu]Cu-DOTA and [⁶⁴Cu]Cu-PSMA-617 [202,203]. An increased liver uptake was seen in patients scanned with [⁶⁴Cu]Cu-PSMA-617 2 h and 22 h post injection as well as in a biodistribution and radiation dosimetry study using [⁶⁴Cu]CuCl₂ [204,205].

6. Potential Radionuclides for the Future Use of PSMA Inhibitors

While radionuclides like ⁶⁸Ga, ¹⁸F, or ¹⁷⁷Lu have become a sort of routine in the use of radiolabeled PSMA inhibitors, other radionuclides are gaining attention due to their specific properties [206,207,208]. For example, ²²⁵Ac can play a pivotal role in targeted alpha therapy (TAT) of prostate cancer [48,51,78,209,210,211,212,213,214,215], but additional preclinical studies and prospective clinical trials need to be performed to secure its safety and efficacy [216,217,218,219,220,221]. The importance of ²²⁵Ac-labeled radiopharmaceuticals is affirmed by efforts to establish GMP-compliant procedures for production and quality control [222,223,224,225,226]. Regulatory aspects on the supply of radionuclides for routine clinical application represent a hurdle that needs to be considered for the translation of new radiopharmaceuticals [227,228].

6.1. ²¹¹Astatine

²¹¹At is a radiohalogen with a half-life of 7.2 h suitable for TAT using small molecules. Several PSMA radioligands have been radiolabeled with ²¹¹At and indicated improved survival rates in PC-3 PIP PSMA-positive tumor bearing mice [229,230]. Although the inclusion of albumin binding moieties increases blood retention and reduces kidney uptake [149], dehalogenation and nephrotoxicity have been the limiting factors for clinical translation. The latest development ²¹¹At-3-Lu [230] showed improved toxicity data for kidney uptake and reduced off-target uptake in organs like the stomach and salivary glands. The infrastructure for the production of ²¹¹At-labeled radiopharmaceuticals is complex though, efforts are made to make them more widely available [231,232], such as the NOAR COST action, which has the goal to create a European Network for Optimized Astatine labeled Radiopharmaceuticals.

6.2. ^64/67Copper

The Donnelly group investigated the bivalent PSMA inhibitor Sar-bisPSMA with the matched pair ⁶⁴Cu/⁶⁷Cu against [⁶⁸Ga]Ga-PSMA-11 and [¹⁷⁷Lu]Lu-PSMA I&T respectively in a LNCaP xenograft model [233,234]. Sar-bisPSMA consists of a sarcophagine based chelator covalently linked to two lysine-ureido-glutamate functional groups, which can be radiolabeled with copper isotopes at room temperature. [⁶⁴Cu]Cu-Sar-bisPSMA showed higher tumor uptake than [⁶⁸Ga]Ga-PSMA-11 1 h after injection (9 ± 1% IA/g vs. 5 ± 1% IA/g). The dose-dependent tumor growth inhibition of [⁶⁷Cu]Cu-Sar-bisPSMA was comparable to [¹⁷⁷Lu]Lu-PSMA I&T (5 MBq: 58 vs. 65%, 30 MBq: 109 vs. 107% against vehicle control). The longer half-life of [⁶⁴Cu]Cu-Sar-bisPSMA allows prospective dosimetry for the use of [⁶⁷Cu]Cu-Sar-bisPSMA. Two clinical trials using [⁶⁴Cu]Cu-Sar-bisPSMA and [⁶⁷Cu]Cu-Sar-bisPSMA have been started in 2021 (NCT04839367/NCT04868604).

Kelly et al. similarly used a sarcophagine chelator (MeCOSar) conjugated to N^ε-(2-(4-iodophenyl)acetyl)lysine as albumin binding group and a PSMA binding motif (RPS-085) [235]. Uptake of [⁶⁴Cu]Cu-RPS-085 in LNCaP tumor bearing mice was constant for the tumor (4 h p.i.: 12.9 ± 1.4% ID/g, 24h p.i.: 8.3 ± 0.8% ID/g, 48 h p.i.: 9.8 ± 1.3% ID/g), while the uptake in the kidneys decreased over time leading to an increased tumor-to-kidney ratio of 6.1 ± 0.8 48 h p.i. The biodistribution was similar when RPS-085 was radiolabeled with ⁶⁷Cu, with a tumor-to-kidney ratio of 3.0 ± 0.5, 96 h p.i.

6.3. ^203/212Lead

²⁰³Pb and ²¹²Pb form an ideal theranostic pair for SPECT imaging and TAT with half-lives of 51.9 h and 10.6 h respectively. The production methods for ^203/212Pb have been improved making them available for potential clinical application [236]. Dos Santos et al. developed four PSMA inhibitors (CA008, CA009, CA011 and CA012) with p-SCN-Bn-TCMC and DO3AM chelators bearing the PSMA-617 binding motif and linker structure [237]. The best ligand PSMA-CA012 radiolabeled with ²⁰³Pb showed similar tumor uptake like [⁶⁸Ga]Ga-PSMA-617 in PSMA-positive C4-2 tumor-bearing mice (8.4 ± 3.7% ID/g vs. 8.5 ± 4.1% ID/g). Two patients have been studied first-in-human with [²⁰³Pb]Pb-PSMA-CA012 and safety dosimetry estimates for [²⁰³Pb]Pb-PSMA-CA012 and [²¹²Pb]Pb-PSMA-CA012 were determined. Stenberg et al. reported the synthesis of NG-001, which has the same chemical structure as CA009 [238]. [²¹²Pb]Pb-NG001 was prepared in situ from a ²²⁴Ra-solution in equilibrium with progeny [239]. [²¹²Pb]Pb-NG001 was evaluated against [²¹²Pb]Pb-PSMA-617 in mice bearing C4-2 tumors. While the tumor uptake was comparable 2 h p.i. (17.61 ± 6.76% ID/g vs. 17.93 ± 2.90% ID/g), kidney uptake was lower for [²¹²Pb]Pb-NG001 (21.07 ± 10.33% ID/g vs. 52.82 ± 26.62% ID/g) [238,239]. Dose-dependent treatment efficiency of [²¹²Pb]Pb-NG001 was confirmed in multicellular C4-2 spheroids and the corresponding mouse model with no long-term toxicity effects [240]. Apart from DOTA and TCMC, pyridine-based cyclen analogs (DOTA-1Py, DOTA-2Py and DOTA-3Py) [236] may be a better chelator alternative for ²¹²Pb and its daughter isotopes for avoiding recoil effects [241].

6.4. ^{149/152/155/161}Terbium

Terbium, also called the “Swiss Army Knife” of Nuclear Medicine with its quadruplet of radioisotopes [242] with suitable properties for imaging with PET (¹⁵²Tb/¹⁴⁹Tb) and SPECT (¹⁵⁵Tb) and endoradiotherapy with alpha (¹⁴⁹Tb) and beta⁻/Auger electrons (¹⁶¹Tb), has been investigated by the group at Paul Scherrer Institute (PSI). Terbium radioisotopes have been used in combination with PSMA-617 in PSMA-positive PC-3 PIP tumor cells. [¹⁵²Tb]Tb-PSMA-617 has been evaluated in vitro and in vivo and successfully applied first-in-human in a patient with metastatic castration-resistant prostate cancer [76]. Suppression of tumor growth has been demonstrated using [¹⁴⁹Tb]Tb-PSMA-617 in TAT in mice bearing PC-3 PIP tumor cells [243]. In a preclinical study, [¹⁶¹Tb]Tb-PSMA-617 was compared against [¹⁷⁷Lu]Lu-PSMA-617 in a PC-3 PIP xenograft model showing the superiority of ¹⁶¹Tb vs. ¹⁷⁷Lu [244,245] opening the way for clinical translation. The additional therapeutic effect of ¹⁶¹Tb due to Auger electrons was confirmed by a computational approach using a microdosimetry model [246]. However, despite these very promising results, the production of terbium radioisotopes for clinical application remains challenging [247,248].

6.5. ⁸⁹Zirconium

Vázquez et al., modified PSMA-617 by replacing DOTA with DFO as chelator making it suitable for radiolabeling with ⁸⁹Zr [249]. ⁸⁹Zr is a positron-emitter with a physical half-life of 3.27 days. The aim of this study was to detect low PSMA-expressing lesions which are negative on PSMA PET scans using short-lived radiotracers, such as [⁶⁸Ga]Ga-PSMA-11 and [¹⁸F]JK-PSMA-7 by increasing the time for internalization of the radioligand. In a LNCaP xenograft model, 24 h after intravenous injection [⁸⁹Zr]Zr-PSMA-DFO showed a higher tumor-to-background ratio than [⁶⁸Ga]Ga-PSMA-11 and [¹⁸F]JK-PSMA-7 after 2 and 4 h. In a small number of patients (8/14) additional lesions were identified with [⁸⁹Zr]Zr-PSMA-DFO after initial scans were negative with [⁶⁸Ga]Ga-PSMA-11 or [¹⁸F]JK-PSMA-7 [250]. An additional [⁸⁹Zr]Zr-PSMA-DFO PET scan at later time points might be beneficial for patients with biochemical recurrence and weak PSMA expression in tumor lesions.

Noor et al. [251] conjugated the Lys-ureido-Glu binding motif to the desferrioxamine B squareamide ester derivative H₃DFOSq, leading to two monovalent and two bivalent DFO-Sq-lysine-ureido-glutamate derivatives with different linker structures to be radiolabeled with ⁸⁹Zr and ⁶⁸Ga. The bivalent ligands performed superior in LNCaP tumor models compared to the monovalent ligands. The bivalent ligand bearing an aromatic hydrophobic linker (H₃DFO-tren-bis(glut-PAMBA-Lys-ureido-Glu)) showed the highest tumor uptake when radiolabeled with ⁸⁹Zr and ⁶⁸Ga compared against [⁶⁸Ga]Ga-PSMA-11 [252] 1 h after injection (9.33 ± 0.33% IA/g vs. 10.8 ± 1.3% IA/g vs. 4.89 ± 1.34% IA/g).

7. Conclusions

The feasibility of theranostics with PSMA-targeting radiopharmaceuticals has been successfully demonstrated in the last decade, upon their radiolabeling using pairs of radionuclides with suitable properties for imaging with PET or SPECT and endoradiotherapy with alpha and beta⁻/Auger electrons. Radioguided and fluorescence-guided surgery should be an additional treatment option for PCa patients, in this direction PSMA ligands for radio-/fluorescence-guided surgery and/or targeted photodynamic therapy are being investigated. In order to further improve the efficiency and efficacy of the existing PSMA radioligands for PCa diagnosis and radionuclide therapy several structural modifications have been suggested for reducing their uptake in off-target tissues or utilizing multitargeting (i.e., albumin, GRPr and integrin α_vβ₃). Addressing additional targets like GRPr with one radiotracer could reduce the number of false-negative findings. New radionuclides in combination with PSMA-binding structures are also being explored, but care should be taken on their radiobiological effects, availability for mass production and cost.