Abstract
Initial studies in patients have demonstrated the suitability of 111In-PSMA-I&T (111In-DOTAGA-(3-iodo-y)-f-k-Sub(KuE)) (PSMA is prostate-specific membrane antigen and I&T is imaging and therapy) for radioguided surgery (RGS) of small metastatic prostate cancer (PCa) soft-tissue lesions. To meet the clinical need for a more cost-effective alternative, the PSMA-I&T–based tracer concept was adapted to 99mTc-labeling chemistry. Two PSMA-I&T–derived inhibitors with all-L-serine- (MAS3) and all-D-serine- (mas3) chelating moieties were evaluated in parallel, and a kit procedure for routine 99mTc labeling was developed. Methods: PSMA affinities (IC50) and internalization kinetics of 99mTc-MAS3-y-nal-k(Sub-KuE) and 99mTc-mas3-y-nal-k(Sub-KuE) (99mTc-PSMA-I&S for imaging and surgery) were determined using LNCaP cells and (125I-BA)KuE as a radioligand and reference standard. In vivo metabolite analyses and biodistribution studies were performed using CD-1 nu/nu and LNCaP tumor–bearing CB-17 severe combined immunodeficiency mice. The pharmacokinetics of 99mTc-PSMA-I&S in humans were investigated in a patient with advanced metastatic PCa via sequential planar whole-body SPECT imaging at 1, 3, 5, and 21 h after injection. Additionally, preoperative SPECT/CT (12 h after injection) and 99mTc-PSMA-I&S–supported RGS (16 h after injection) were performed in 1 PCa patient with proven iliac and inguinal lymph node metastases. Results: A robust and reliable kit-labeling procedure was established, allowing the preparation of 99mTc-MAS3-y-nal-k(Sub-KuE) and 99mTc-PSMA-I&S in consistently high radiochemical yield and purity (≥98%, n > 50 preparations). Because of its improved internalization efficiency and superior in vivo stability, 99mTc-PSMA-I&S was selected for further in vivo evaluation. Compared with 111In-PSMA-I&T, 99mTc-PSMA-I&S showed delayed clearance kinetics but identical uptake in PSMA-positive tissues in the LNCaP xenograft model (1 h after injection). In exemplary PCa patients, a relatively slow whole-body clearance of 99mTc-PSMA-I&S was observed due to high plasma protein binding (94%) of the tracer. This, however, promoted efficient tracer uptake in PCa lesions over time and led to steadily increasing lesion-to-background ratios up to 21 h after injection. Preoperative SPECT/CT showed a high 99mTc-PSMA-I&S uptake in all suspect lesions identified in previous 68Ga-HBED-CC-Ahx-KuE (68Ga-HBED-CC-PSMA) PET/CT, allowing for their successful intraoperative detection and resection during first-in-human RGS. Conclusion: Because of a straightforward and reliable kit production, 99mTc-PSMA-I&S represents a cost-effective, readily available alternative to 111In-PSMA-I&T. Initial patient data indicate its comparable or even superior performance as a probe for PSMA-targeted RGS and also hint toward the unexpected potential of 99mTc-PSMA-I&S as a SPECT imaging agent.
Recently, prostate-specific membrane antigen (PSMA) has emerged as one of the most extensively investigated and exploited targets for molecular imaging and radioligand therapy of prostate cancer (PCa). Because of its strong upregulation in PCa and low basal expression in nonprostatic tissues as well as the direct correlation between PSMA expression levels and androgen independence, metastasis, and PCa progression (1,2), PSMA represents a highly valuable molecular marker in PCa.
Therefore, intense efforts have been directed toward the development of PSMA-targeted probes for a variety of clinical applications ranging from diagnostic imaging using SPECT, PET, MRI, or optical methods toward innovative theranostic and therapeutic concepts (3–6).
With respect to clinical diagnostic imaging, the field was pioneered by small urea-based tracers such as 99mTc-MIP-1404 and 99mTc-MIP-1405 for SPECT (7,8) and by 68Ga-HBED-CC-Ahx-KuE (68Ga-HBED-CC-PSMA) (9–11) or the 18F-labeled analogs 18F-DCFBC (12,13) and 18F-DCFPyl (14,15) for PET. Their use for PET/CT or PET/MRI hybrid imaging has been shown to allow detection and characterization of primary and recurrent metastatic PCa, even at low prostate-specific antigen levels, with higher sensitivity, specificity, and accuracy than conventional imaging methods (16–18).
Furthermore, several alternative theranostic approaches have been realized recently, providing versatile molecular platforms for labeling with diagnostic (123I/124I, 68Ga) and therapeutic radionuclides (131I, 177Lu). Initial data on endoradiotherapeutic applications of 131I-MIP-1095 (19), 177Lu-DKFZ-617 (20,21), and 177Lu-PSMA-I&T (I&T is imaging and therapy) (22,23) in patients with metastatic PCa demonstrate comparable and highly promising molecular and morphologic treatment responses for all 3 compounds (3). In addition, the concept of PSMA-targeted therapy can be expanded beyond endoradiotherapy toward radioguided surgery (RGS), as recently demonstrated by the introduction of 111In-PSMA-I&T (24,25). Preoperative injection of 111In-PSMA-I&T was shown to substantially facilitate the intraoperative detection and resection of even small and atypically localized PSMA-positive metastatic lymph node deposits. In a small cohort of patients with primary metastasized or early recurrent PCa, 111In-PSMA-I&T–supported RGS allowed quantitative resection of all suspect lesions previously identified in 68Ga-HBED-CC-PSMA PET/CT. Furthermore, 111In-PSMA-I&T also showed reasonable in vivo performance as a PSMA-targeted imaging probe in pretherapeutic SPECT/CT imaging at 4 h after injection (24,25). However, its suboptimal nuclear properties, the high cost, and the limited availability of 111InCl3 restrict the routine clinical application of 111In-PSMA-I&T for PSMA-targeted RGS. To circumvent these issues, the development of a corresponding 99mTc-labeled analog seemed the conclusive next step. Here, major focus was directed toward the selection of a 99mTc-labeling strategy that ensured fast and robust radiolabeling and a straightforward translation into a kit formulation for clinical application.
To adapt the PSMA-I&T–based theranostic concept to the requirements of 99mTc labeling, the DOTAGA-chelator in PSMA-I&T was replaced by mercaptoacetyl triserine, named MAS3 (26,27). Furthermore, the 3-iodo-d-Tyr- d-Phe-sequence in the linker unit was replaced by d-Tyr- d-2-Nal to enhance interaction of the peptidic linker unit with a remote arene binding site (Fig. 1). On the basis of earlier work that had demonstrated the metabolic instability of the peptidic linker in DOTAGA-FFK-Sub-KuE (28), we hypothesized that the all-L-amino acid chelator MAS3 (2-mercaptoacetyl-Ser-Ser-Ser-) might also be susceptible toward proteolytic degradation. To investigate this hypothesis, the corresponding mas3 (2-mercaptoacetyl-d-Ser-d-Ser-d-Ser-) analog (PSMA-I&S [for imaging and surgery]) was also synthesized and evaluated in parallel.
Chemical structures of 111In-PSMA-I&T and 99mTc-MAS3/mas3-y-nal-k-Sub-KuE.
MATERIALS AND METHODS
General
All animal experiments were conducted in accordance with the German Animal Welfare Act (Deutsches Tierschutzgesetz, approval no. 55.2-1-54-2532-71-13). All human studies were approved by the institutional review boards of the participating medical institutions. Patients provided signed informed consent.
Synthesis and Radiolabeling
The synthesis of MAS3/mas3-y-naI-k(Sub-KuE) was performed in analogy to a previously published protocol (28), with minor modifications (Fig. 1). A detailed description of the 99mTc-labeling protocol is supplied in the supplemental data (available at http://jnm.snmjournals.org). The radioiodinated reference ligand (125I-BA)KuE ((S)-1-carboxy-5-(4-(-125I-iodo-benzamido)pentyl)carbamoyl)-l-glutamic acid) was prepared as described previously (28).
Lipophilicity and Plasma Protein Binding
The n-octanol/phosphate-buffered saline (O/PBS) partition coefficients of 99mTc-PSMA-I&S and 99mTc-MAS3-y-naI-k(Sub-KuE) were determined as described using a shake-flask method (28). Log PO/PBS values were calculated from the means of 6 separate determinations. The fraction of plasma protein–bound tracer was determined by incubation of fresh human plasma samples with 99mTc-PSMA-I&S and 99mTc-MAS3-y-naI-k(Sub-KuE), respectively (30 min, 37°C), and subsequent ultracentrifugation in VWR Centrifugal Filters (PES, 30K; VWR). The fraction of free 99mTc inhibitor in plasma was defined as the ratio between the activity in the ultrafiltrate and in the unfiltered plasma sample. Values for plasma protein binding were corrected for nonspecific binding to the filter material.
In Vitro Studies
PSMA-positive LNCaP cells (300265; CLS) were cultivated in Dulbecco modified Eagle medium/Nutrition Mix F-12 with Glutamax-I (1:1) (Invitrogen, Life Technologies) supplemented with 10% fetal calf serum and were maintained at 37°C in a 5% CO2/humidified air atmosphere. For IC50 (PSMA affinity) determination, 150,000 cells/well were seeded in conventional 24-well plates 1 d before the experiment. For internalization studies, 125,000 cells/well and PLL-coated 24-well plates were used.
IC50 was determined in a competitive binding assay using LNCaP cells and (125I-BA)KuE as the radioligand (28). Data represent mean ± SD of 3 or more separate determinations.
PSMA-specific ligand internalization kinetics were determined by incubation of LNCaP cells (37°C; 5, 15, 30, and 60 min, respectively) with the respective radioligands (0.2 nM) in the absence (total internalization) and presence (nonspecific internalization) of 10 μM 2-PMPA (2-(phosphonomethyl)pentane-1,5-dioic acid). Data were corrected for nonspecific internalization and normalized to the specific internalization observed for the reference compound (125I-BA)KuE in a parallel experiment (22). Data are mean ± SD (n = 3).
Metabolite Analysis
Approximately 20 MBq of the 99mTc-labeled tracers (0.5–0.6 nmol; specific activity, 36 GBq/μmol) were injected into the tail vein of CD-1 nu/nu mice. At 1 h after injection, animals were sacrificed; blood and urine were collected; and kidneys were dissected. Sample preparation and reversed-phase high-performance liquid chromatography (RP-HPLC) analysis were performed in analogy to a previous protocol (28) using a Chromolith Performance RP-18e column (CS-Chromatographie GmbH; 100 × 4.6 mm; flow rate, 3 mL/min; gradient, 3% B for 3 min, 3%–95% B in 6 min, 95% B for 3 min; solvent A, 0.1% trifluoroacetic acid in water; solvent B, 0.1% trifluoroacetic acid in acetonitrile).
In Vivo Biodistribution Studies
To induce LNCaP tumor growth, CB17 severe combined immunodeficiency mice (6–8 wk, male, Charles River Laboratories) were inoculated subcutaneously with approximately 1 × 107 LNCaP cells in 100 μL of serum-free medium and 100 μL of Matrigel (BD Biosciences). After an average of 4–6 wk, tumor size reached 4–8 mm in diameter, and the animals were used for biodistribution studies.
Biodistribution
About 3–4 MBq of 99mTc-PSMA-I&S (0.1 nmol) or 1.5 MBq of 111In-PSMA-I&T (0.2 nmol) (24) were injected into the tail vein of the animals (n = 5) under isoflurane anesthesia. For competition experiments, 2-PMPA (1 μmol = 226 μg/mouse) was coinjected (n = 3). Animals were sacrificed at 1 h after injection, the organs of interest were dissected, and the activity in the weighed tissues samples was quantified using a γ-counter.
99mTc-PSMA-I&S and RGS in Patients
Diagnostic 68Ga-HBED-CC-PSMA PET/CT imaging of both PCa patients in this study was performed as described recently (10).
In 1 patient, whole-body scintigraphy and SPECT/CT were performed 1, 3, 5, and 21 h after intravenous administration of 497 MBq of 99mTc-PSMA-I&S on a Siemens Symbia T 6. Planar whole-body images were acquired with a continuous table feed of 10 cm/min, immediately followed by SPECT/CT acquisition. The SPECT acquisition (64 × 64 matrix, 64 frames, 30 s/frame) was performed using 6 angular steps in a 20-s time frame. For CT (130 kV, 15 mAs), 5-mm slices were obtained.
First-in-human RGS was performed in 1 patient with primary metastatic PCa 16 h after intravenous injection of 607 MBq of 99mTc-PSMA-I&S. After preoperative SPECT/CT 12 h after injection, metastatic lymph nodes were detected intraoperatively using a γ-probe (Crystal Probe CXS-SG603; Crystal Photonics).
RESULTS
Chemical Synthesis
The synthesis of the novel PSMA-I&T–derived ligands is summarized in Figure 2. Because of a greater ease of synthesis and better synthetic availability, Sub(OPfp)2 (suberic acid bis-pentafluorophenyl ester) was used to prepare the protected linker-conjugated inhibitor component (OtBu)K(Sub-OPfp)uE(OtBu)2 (2) in 68% yield based on 1. On reaction with the respective protected peptidic linker units 3 and 4 that had been prepared via standard solid-phase peptide synthesis in sufficient purity to allow immediate use for further reaction, the conjugates were deprotected and purified using preparative RP-HPLC. The final labeling precursors 5 and 6 were obtained in 71% and 63% yields, respectively, based on chelator-conjugated peptide spacer. The natRe complexes of 5 and 6 were synthesized under weakly acidic conditions using KReO4 and SnCl2 (90°C, 1 h). To remove excess salts, Re-5 and Re-6 were purified using RP-HPLC. The identity of all final products was confirmed by electrospray ionization mass spectrometry.
Synthesis of MAS3-y-nal-k-Sub-KuE (5) and mas3-y-nal-k-Sub-KuE (6, PSMA-I&S). (A) Piperidine (20%) in NMP, Fmoc-D-2NaI-OH, HOBt, TBTU, DIPEA, [NMP]. (B) Piperidine (20%) in NMP, Fmoc-D-Tyr(tBu)-OH, HOBt, TBTU, DIPEA, [NMP]. (C) Piperidine (20%) in NMP, Fmoc-D-Ser(tBu)-OH, HOBt, TBTU, DIPEA, [NMP]. (D) Piperidine (20%) in NMP, S-Trityl-mercaptoacetic acid, HOBt, TBTU, DIPEA, [NMP]. (E) DCM/TFE/acetic acid (6/3/1, (v/v)). (F) Piperidine (20%) in NMP, Fmoc-L-Ser(tBu)-OH, HOBt, TBTU, DIPEA, [NMP]. (G) Sub(OPfp)2, TEA, [DMF]. (H) 3, TEA, [DMF]. (I) TFA. (J) 4, TEA, [DMF].
99mTc Labeling
For initial 99mTc-labeling experiments, a standard wet chemistry MAG3-labeling procedure (29) was adapted to the requirements for the production of high-specific-activity radiopharmaceuticals using only 20–30 nmol of labeling precursors 5 and 6, respectively. Both for 99mTc-5 (99mTc-PSMA-I&S) and 99mTc-6, no free 99mTc-pertechnetate was detected in the reaction mixture after heating the respective precursor and 99mTc-pertechnetate (1,000–1,200 MBq) for 20 min to 90°C in the presence of stannous chloride, ascorbic acid, tartrate, and ammonium acetate (pH 7.5–8). However, substantial amounts (10%–30% of added activity) of colloidal 99mTc species were detected in the respective preparations, necessitating a cartridge purification of 99mTc-PSMA-I&S and 99mTc-6, which were obtained in specific activities of 25–37 GBq/μmol.
Subsequently, the composition of the initial wet chemistry labeling mixture was adapted to the requirements for kit formulation. Thus, the volatile ammonium acetate buffer (pH 8) was replaced by phosphate buffer (pH 8), and single-dose freeze-dried kits containing 25 nmol of PSMA-I&S were prepared. As observed in the initial labeling studies, the radiochemical yield of 99mTc-PSMA-I&S was also limited to 65%–87% (n = 5) in the kit preparations because of the formation of colloidal 99mTc species. Neither increasing the amount of labeling precursor nor adding a freshly prepared SnCl2 solution to the kit immediately before the addition of 99mTc-pertechnetate nor using a higher amount of the reducing agent (50 μg) SnCl2 or addition of ethanol to the kit reaction mixture (30) repressed 99mTc-colloid formation, whereas adjustment of the tartrate content in the lyophilized kits allowed the production of 99mTc-PSMA-I&S in consistently high radiochemical yield and radiochemical purity of 99% or greater (n > 50 kit preparations), with negligible formation of 99mTc colloid (≤1% of added 99mTc-activity). The kit preparation of 99mTc-PSMA-I&S was found to be highly tolerant toward concentration and volume changes (1–10 mL reaction volume), affording 99mTc-PSMA-I&S in improved specific activities of 44–52 GBq/μmol.
On the basis of these findings, the cartridge purification step after 99mTc labeling can safely be omitted. The synthesis is completed by dilution with sterile isotonic saline and sterile filtration.
In Vitro Evaluation
The IC50 of the precursors 5 and 6 and of the corresponding natRe complexes of 5 and 6 were determined in a competitive binding assay using LNCaP human PCa cells and (125I-BA)KuE (31) as the radioligand (0.2 nM; Table 1).
Both in the free and in the Re-complexed form, the novel MAS3/mas3-y-naI-k(Sub-KuE) analogs exhibit similar IC50, indicating a negligible influence of the d- versus l-conformation of the serine residues in the mas3/MAS3 chelators as well as of the amino acid composition of the peptidic linker (y-nal-k vs. (3-iodo-y)-f-k in PSMA-I&T) on IC50. Compared with the reference compound natIn-PSMA-I&T, both Re-PSMA-I&S (Re-5) and Re-6 show slightly reduced affinities toward PSMA.
However, marked differences in the internalization efficiencies of 99mTc-PSMA-I&S (99mTc-5) (Fig. 3) and 99mTc-MAS3-y-naI-k(Sub-KuE) (99mTc-6) were observed. Compared with 99mTc-6, 99mTc-PSMA-I&S showed enhanced internalization into LNCaP cells, which almost reached the values obtained for 111In-PSMA-I&T, whereas the internalization of 99mTc-6 was substantially lower and comparable to that of 177Lu-PSMA-I&T.
(Left) Kinetics of PSMA-mediated total cellular uptake and internalization of 99mTc-PSMA-I&S (0.2 nM) into LNCaP cells (37°C) (mean ± SD [n = 3]). (Right) Kinetics of PSMA-mediated internalization of 99mTc-PSMA-I&S (0.2 nM) and reference (125I-BA)KuE (0.2 nM) into LNCaP cells (37°C) (mean ± SD [n = 3]).
Lipophilicity and Plasma Protein Binding
The lipophilicities of 99mTc-PSMA-I&S and 99mTc-6 are summarized in Table 1. Compared with the highly hydrophilic PSMA-I&T analogs, lipophilicity of the novel 99mTc-labeled compounds, albeit still low, was increased by an order of magnitude.
The fraction of protein-bound tracer (human plasma) was found to be 94% and 95% for 99mTc-PSMA-I&S and 99mTc-6, respectively, which was considerably higher than the value observed for 111In-PSMA-I&T (83%) (24).
In Vivo Metabolite Analysis
The in vivo metabolic stability of 99mTc-PSMA-I&S and 99mTc-6 was investigated in CD-1 mice (Fig. 4). As expected, no in vivo degradation was observed for the stabilized analog 99mTc-PSMA-I&S, whereas substantial radiometabolite formation was observed for 99mTc-6 containing the L-amino acid MAS3-chelator. In blood and urine, the percentage of intact 99mTc-6 after 1 h was only 52% and 88%, respectively, whereas no tracer degradation was observed in the kidneys, suggesting predominant metabolism of 99mTc-6 in plasma and fast renal excretion of the radiometabolite.
In vivo stability of 99mTc-PSMA-I&S and 99mTc-6 in CD-1 nu/nu mice. Radio–high-performance liquid chromatograms of intact tracer (before injection) and cell-free blood, urine, and kidney homogenate samples collected 1 h after injection of respective radioligands.
Biodistribution
Given the superior metabolic stability of 99mTc-PSMA-I&S, only this tracer was further investigated in an in vivo biodistribution study (Table 2). As expected from its pronounced plasma protein binding, 99mTc-PSMA-I&S showed delayed blood clearance and thus increased blood and background activity levels compared with 111In-PSMA I&T at 1 h after injection. The reduced hydrophilicity of 99mTc-PSMA-I&S led to enhanced tracer uptake in the liver and intestines, indicating an increased contribution of hepatobiliary clearance to overall tracer excretion. Interestingly, tracer accumulation in PSMA-expressing tissues, that is, the spleen, kidneys, and LNCaP xenograft, was identical for both 99mTc-PSMA-I&S and 111In-PSMA-I&T, underlining the in vitro results (Table 1) that had already shown comparable PSMA targeting efficiencies for both compounds. PSMA specificity of 99mTc-PSMA-I&S accumulation was confirmed by a competition study using a 1,000-fold molar excess of the potent PSMA inhibitor PMPA (Table 2). Here, PSMA-mediated 99mTc-PSMA-I&S uptake in the spleen, kidneys, and tumor was reduced to 3%, 5%, and 22%, respectively, of the tissue accumulation under tracer-only conditions.
Biodistribution of 99mTc-PSMA-I&S and Reference 111In-PSMA-I&T (24) in LNCaP Tumor–Bearing CB-17 Severe Combined Immunodeficiency Mice at 1 Hour After Injection
First Patient Application of 99mTc-PSMA-I&S for SPECT Imaging and RGS
On the basis of these in vivo data, a first-in-human RGS study was performed. To investigate tracer kinetics and to identify suitable time points for preoperative SPECT imaging and subsequent RGS, a 73-y-old patient with metastatic castration-resistant PCa was sequentially imaged over a total of 21 h (Fig. 5). Analogous to a prior 68Ga-HBED-CC PSMA PET/CT scan (Fig. 5A), ventral and dorsal 99mTc-PSMA-I&S whole-body planar scintigraphies (Figs. 5B–5E) revealed diffuse bone and lymph node metastases as early as 1 h after injection (Fig. 5B). As expected from the preclinical data, 99mTc-PSMA-I&S showed delayed whole-body clearance and efficient accumulation in PSMA-expressing tissue (Fig. 5D). Although 99mTc-PSMA-I&S showed some diffuse uptake in the liver and gastrointestinal tract, most of the tracer was cleared from these and other organs within the observation period. In contrast, 99mTc-PSMA-I&S uptake in parotid and salivary glands and PSMA-mediated uptake in kidneys as well as bone and lymph node metastases increased between 1 and 3 h after injection and remained persistently high. These combined effects led to gradually increasing lesion-to-background ratios up to 21 h after injection (Fig. 5E).
68Ga-HBED-CC PSMA PET (A; maximum-intensity projection, 1 h after injection) and 99mTc-PSMA-I&S whole-body planar scintigraphy (B–E) were performed in a PCa patient with metastatic, hormone-refractory disease. (B–E) Whole-body planar scintigraphy at 1 (B), 3 (C), 5 (D), and 21 h (E) after injection of approximately 500 MBq of 99mTc-PSMA-I&S.
Another patient (72 y), presenting with histologically confirmed primary PCa (Gleason score, 7b; pT2v, cNx, Mx; initial prostate-specific antigen, 13 ng/mL) initially underwent 68Ga-HBED-CC-PSMA PET/MR for pretherapeutic staging. PET/MR showed iliac and atypically localized inguinal lymph node metastases with intense 68Ga-HBED-CC-PSMA uptake (Fig. 6). The patient was subsequently scheduled for RGS. 99mTc-PSMA-I&S SPECT/CT images at 12 h after injection revealed intense PSMA uptake in all previously identified lesions (68Ga-HBED-CC-PSMA PET/MR), facilitating intraoperative identification of the suspect lymph nodes. The presence of PSMA-positive tumor tissue in the resected specimens was confirmed histopathologically after RGS.
Preoperative 68Ga-HBED-CC PSMA PET/MR (A–D) and 99mTc-PSMA-I&S SPECT/CT (E–H). 68Ga-HBED-CC PSMA PET/MR shows iliac (A and B) and inguinal lymph node metastases (C and D). 99mTc-PSMA-I&S SPECT/CT at 12 h after injection reveals intense tracer uptake in same lymph nodes.
DISCUSSION
The recent introduction of 111In-PSMA-I&T for PSMA-targeted RGS and optional preoperative SPECT imaging has further widened the scope of current PSMA-directed theranostic concepts. However, the inherent limitations associated with 111In as a radionuclide prevent the broader clinical use of 111In-PSMA-I&T beyond proof-of-concept studies in small patient cohorts. Thus, the aim of this study was to adapt the general tracer concept to the requirements of 99mTc chemistry.
The PSMA-I&T scaffold is tolerant toward structural modifications in the N-terminal region of the peptidic linker unit without pronounced effects on IC50 (22,28). Therefore, the different well-established 99mTc-labeling methodologies such as hydrazinonicotinic acid–(HYNIC), 99mTc(CO)3+-, and MAG3-based strategies seemed equally well suited for the design of novel 99mTc-labeled PSMA inhibitors. In addition, robust and reliable kit procedures have been developed (32–34) for these approaches, and the choice of suitable chelators or coligand systems allows fine tuning of ligand pharmacokinetics via adjustment of hydrophilicity (27). MAG3-based 99mTc-labeling approaches, however, have the advantage of neither requiring additional coligands for 99mTc complexation (as opposed to HYNIC-functionalized precursors) nor necessitating the synthesis of complex chelator systems (as opposed to 99mTc(CO)3+ chelation). The use of the hydrophilic MAG3-analog MAS3 (mercaptoacetyl-seryl-seryl-serine) leads to the formation of a well-defined, hydrophilic 99mTc complex while maintaining the ease and efficiency of the MAG3 99mTc-labeling procedure (26).
These features prompted the integration of MAS3 into our PSMA-I&T–based tracer design (Fig. 1). To the best of our knowledge, the potential influence of MAS3 stereochemistry on 99mTc labeling and stability of the corresponding 99mTc radiopharmaceuticals has not been investigated so far. Thus, to close this gap, both the PSMA inhibitor bearing the all-L-serine- (MAS3) and the all-D-serine- (mas3) chelator (99mTc-6 and 99mTc-PSMA-I&S [99mTc-5], respectively) were evaluated in parallel.
No detectable influence of chelator stereochemistry on the outcome of the 99mTc-labeling reaction was observed, because the formation of the 99mTcO-MAS3/mas3 complex should be independent from the spatial orientation of the serine side chains. As hypothesized, however, on the basis of the intrinsic susceptibility of L-amino acid peptides toward in vivo degradation by endopeptidases, the MAS3 analog 99mTc-6 showed substantially decreased in vivo stability compared with the mas3-derivative 99mTc-PSMA-I&S, for which only intact tracer was detected in blood, urine, and kidneys at 1 h after injection (Fig. 4). In the case of 99mTc-6, rapid formation of a hydrophilic radiometabolite in blood was observed; this metabolite was not detected in the kidney homogenate, but in the urine, indicating efficient renal clearance. Although the identity of the radiometabolite was not further investigated, it seems probable that enzymatic cleavage of the scissile bond between d-Tyr and the first l-Ser residue of MAS3 leads to the formation of free 99mTc-MAS3, which—in analogy to MAG3 in renal scintigraphy—is then readily filtered into the urine without nonspecific retention in kidneys.
The present study was focused on the development of a suitable 99mTc-labeled probe for PSMA-targeted RGS. Here, a high in vivo stability represents a major prerequisite. In the clinical setting, RGS is usually performed on the day after tracer injection for practical reasons (35). Its success primarily relies on high lesion-to-background contrast at the time of surgery rather than fast tracer kinetics. The high in vivo stability of 99mTc-PSMA-I&S ensures the prolonged availability of intact tracer in the circulation, which can be expected to lead to progressively increasing activity accumulation in PSMA-expressing lesions over time.
This effect may be further amplified by the higher plasma protein binding displayed by 99mTc-PSMA-I&S (Table 1). In human plasma, approximately 94% of 99mTc-PSMA-I&S is protein-bound, whereas this fraction amounts to only 83% for 111In-PSMA-I&T. This is mirrored in the biodistribution data for these 2 compounds (Table 2), where 99mTc-PSMA-I&S shows a 7-fold-higher blood activity concentration at 1 h after injection than 111In-PSMA-I&T. As early as 1 h after injection, however, this was not immediately reflected in enhanced targeting of 99mTc-PSMA-I&S to PSMA-overexpressing tissues such as kidney and tumor. Both 99mTc-PSMA-I&S and 111In-PSMA-I&T showed identical uptake values in these tissues (Table 2). As demonstrated by the blocking experiment (Table 2), however, 99mTc-PSMA-I&S uptake in kidney and tumor was highly PSMA-specific.
On the basis of the promising in vivo data obtained for 99mTc-PSMA-I&S, sequential planar imaging (1–21 h after injection, Fig. 5) was performed in 1 exemplary patient to establish the pharmacokinetics of 99mTc-PSMA-I&S in humans and to identify a suitable time window for RGS, where optimal lesion-to-background contrast is achieved, but intraoperative activity detection is not yet compromised by the decay of 99mTc.
As anticipated, whole-body clearance of 99mTc-PSMA-I&S was comparably slow, leading to relatively late background clearance, especially in the abdominal region. In accordance with our assumptions, however, tracer accumulation in the previously identified PCa lesions steadily increased over time as a result of the prolonged availability of the intact tracer in the blood and its internalization into the PSMA-expressing tumor cells. At later points 5 h or more after injection, excellent lesion-to-background ratios were obtained because of the synergistic effect of persistent 99mTc-PSMA-I&S uptake in tumor tissue and continuing clearance of background activity.
On the basis of these data, 99mTc-PSMA-I&S–supported RGS was performed at 16 h after injection in an exemplary patient. As demonstrated in Figure 6, preoperative 99mTc-PSMA-I&S SPECT/CT (12 h after injection) revealed high-focal-activity accumulation in several lymph node metastases, allowing for exact intraoperative identification and resection during RGS. It is important to note that although preliminary in nature, these first-in-human studies suggest improved performance of 99mTc-PSMA-I&S compared with 111In-PSMA-I&T, especially with respect to absolute tracer uptake in PCa lesions and, consequently, imaging contrast in preoperative SPECT.
Of course, the primary focus in the development of 99mTc-PSMA-I&S was its adaptation to the requirements of RGS. Preoperative SPECT imaging was primarily performed to confirm sufficient tracer uptake in all suspect lesions previously identified via 68Ga-HBED-CC-PSMA PET and thus to ensure their detectability during RGS. Surprisingly, however, the quality of planar and SPECT images obtained with 99mTc-PSMA-I&S compares well to images obtained with alternative highly promising PSMA-targeted SPECT imaging agents such as 99mTc-MIP-1404 (99mTc-trofolastat) (8). Although the latter had displayed fundamentally different pharmacokinetics in LNCaP xenograft–bearing nude mice, that is, fast background clearance and slightly increased tumor accumulation (36), as well as substantially lower plasma protein binding (32%) (34), suitable imaging contrast in PCa patients was obtained only at time points of 4 h or more after injection because of considerable hepatic and intestinal tracer accumulation (8). In this context, a further evaluation of 99mTc-PSMA-I&S with emphasis on its suitability as a SPECT imaging agent seems highly recommendable. If our preliminary data are confirmed, they may even support the notion of first-line diagnosis of metastasized PCa via SPECT imaging in centers at which PET is not available. It remains questionable, however, whether the late imaging time points required for high-contrast PCa imaging using the currently available 99mTc-labeled PSMA ligands are compatible with the patient workflow in imaging centers.
With respect to availability and ease of tracer preparation, 99mTc-PSMA-I&S is fully compatible with everyday clinical workflow. As a result of this study, a robust and reliable freeze-dried kit is now available for routine production of 99mTc-PSMA-I&S, facilitating the distribution and on-site production of this radiopharmaceutical for clinical applications in urology (RGS) and nuclear medicine (SPECT).
CONCLUSION
On the basis of the preclinical and first patient data obtained in this study, 99mTc-PSMA-I&S is a superior substitute for 111In-PSMA-I&T for PSMA-targeted RGS. Its high value for the intraoperative detection of small metastatic lesions in PCa patients during RGS is supported by the results of currently ongoing RGS studies in more than 40 patients and may, alongside continuous advances in γ-probe technology, help to support the progressive integration of the concept of PSMA-targeted RGS into the clinic.
DISCLOSURE
The current study was financially supported by the Deutsche Forschungsgemeinschaft (SFB824; subproject Z1 and A8). No other potential conflict of interest relevant to this article was reported.
Acknowledgments
We thank Alexandra Bartel, Simone Loher, and Sven Hintze for their excellent technical assistance. Furthermore, we thank Gregor Weirich for performing the histophathologic analysis of the resected tissues.
Footnotes
↵* Contributed equally to this work.
Published online Sep. 15, 2016.
- © 2017 by the Society of Nuclear Medicine and Molecular Imaging.
REFERENCES
- Received for publication May 31, 2016.
- Accepted for publication August 1, 2016.
Jump to section
Related Articles
Cited By...
- Receptor-Targeted Peptide Conjugates Based on Diphosphines Enable Preparation of 99mTc and 188Re Theranostic Agents for Prostate Cancer
- Strong Correlation Between SUVmax on PSMA PET/CT and Numeric Drop-In {gamma}-Probe Signal for Intraoperative Identification of Prostate Cancer Lesions
- Prostate-Specific Membrane Antigen-Targeted Radioguided Pelvic Lymph Node Dissection in Newly Diagnosed Prostate Cancer Patients with a Suspicion of Locoregional Lymph Node Metastases: The DETECT Trial
- Delayed Imaging Improves Lesion Detectability in [99mTc]Tc-PSMA-I&S SPECT/CT in Recurrent Prostate Cancer
- A 2022 International Survey on the Status of Prostate Cancer Theranostics
- Occupational Radiation Exposure of Radiopharmacy, Nuclear Medicine, and Surgical Personnel During Use of [99mTc]Tc-PSMA-I&S for Prostate Cancer Surgery
- Radiation Dosimetry of 99mTc-PSMA I&S: A Single-Center Prospective Study
- Can Intraoperative Fluorescence Imaging Identify All Lesions While the Road Map Created by Preoperative Nuclear Imaging Is Masked?
- Hybrid Tracers Based on Cyanine Backbones Targeting Prostate-Specific Membrane Antigen: Tuning Pharmacokinetic Properties and Exploring Dye-Protein Interaction
- Trending: Radioactive and Fluorescent Bimodal/Hybrid Tracers as Multiplexing Solutions for Surgical Guidance
- Prostate-Specific Membrane Antigen-Guided Surgery
- Future of Theranostics: An Outlook on Precision Oncology in Nuclear Medicine
- Entering the Era of Molecularly Targeted Precision Surgery in Recurrent Prostate Cancer
- Synthesis and Preclinical Characterization of the PSMA-Targeted Hybrid Tracer PSMA-I&F for Nuclear and Fluorescence Imaging of Prostate Cancer
- PSMA Ligands for PET Imaging of Prostate Cancer
- Prostate-Specific Membrane Antigen Ligands for Imaging and Therapy
- Glu-Ureido-Based Inhibitors of Prostate-Specific Membrane Antigen: Lessons Learned During the Development of a Novel Class of Low-Molecular-Weight Theranostic Radiotracers
- Phase 2 Study of 99mTc-Trofolastat SPECT/CT to Identify and Localize Prostate Cancer in Intermediate- and High-Risk Patients Undergoing Radical Prostatectomy and Extended Pelvic LN Dissection