[111In]In/[177Lu]Lu-AAZTA5-LM4 SST2R-Antagonists in Cancer Theranostics: From Preclinical Testing to First Patient Results

Aiming to expand the application of the SST2R-antagonist LM4 (DPhe-c[DCys-4Pal-DAph(Cbm)-Lys-Thr-Cys]-DTyr-NH2) beyond [68Ga]Ga-DATA5m-LM4 PET/CT (DATA5m, (6-pentanoic acid)-6-(amino)methy-1,4-diazepinetriacetate), we now introduce AAZTA5-LM4 (AAZTA5, 1,4-bis(carboxymethyl)-6-[bis(carboxymethyl)]amino-6-[pentanoic-acid]perhydro-1,4-diazepine), allowing for the convenient coordination of trivalent radiometals of clinical interest, such as In-111 (for SPECT/CT) or Lu-177 (for radionuclide therapy). After labeling, the preclinical profiles of [111In]In-AAZTA5-LM4 and [177Lu]Lu-AAZTA5-LM4 were compared in HEK293-SST2R cells and double HEK293-SST2R/wtHEK293 tumor-bearing mice using [111In]In-DOTA-LM3 and [177Lu]Lu-DOTA-LM3 as references. The biodistribution of [177Lu]Lu-AAZTA5-LM4 was additionally studied for the first time in a NET patient. Both [111In]In-AAZTA5-LM4 and [177Lu]Lu-AAZTA5-LM4 displayed high and selective targeting of the HEK293-SST2R tumors in mice and fast background clearance via the kidneys and the urinary system. This pattern was reproduced for [177Lu]Lu-AAZTA5-LM4 in the patient according to SPECT/CT results in a monitoring time span of 4–72 h pi. In view of the above, we may conclude that [177Lu]Lu-AAZTA5-LM4 shows promise as a therapeutic radiopharmaceutical candidate for SST2R-expressing human NETs, based on previous [68Ga]Ga-DATA5m-LM4 PET/CT, but further studies are needed to fully assess its clinical value. Furthermore, [111In]In-AAZTA5-LM4 SPECT/CT may represent a legitimate alternative diagnostic option in cases where PET/CT is not available.


Introduction
The successful advent of theranostic octreotide analogs in the clinic to combat somatostatin subtype 2 receptor (SST 2 R)-expressing neuroendocrine tumors (NETs), such as the recently approved theranostic pair [ 68 Ga]Ga/[ 177 Lu]Lu-DOTA-TATE, has paved the way for new exciting developments in the field of receptor-targeted diagnostic imaging and radionuclide therapy of human tumors [1][2][3][4][5][6][7]. In this concept, radionuclide therapy is based on previous PET/CT to identify patients eligible to receive it, perform dosimetry and optimize therapeutic schemes. Furthermore, post-therapy PET/CT is applied to assess therapeutic responses and monitor disease progress in an approach abiding to modern personalized medicine principles [8,9]. A recent breakthrough in the field of NET theranostics has been the shift of paradigm from radiolabeled SST 2 R-agonists to antagonists [10][11][12][13][14][15][16][17]. Despite their inability to internalize in cancer cells, radiolabeled SST 2 R-antagonists were shown to bind to more receptor sites on cancer cells, comprising both active and non-active receptor populations, resulting in higher tumor uptake in animal models and patients. Moreover, background clearance turned out to be favorably faster for antagonists for reasons not yet fully understood. Hence, radiolabeled SST 2 R-antagonists are expected to dynamically enter the clinic in the future with the development of new radiopharmaceuticals of higher diagnostic contrast and therapeutic index and potentially also broader clinical indications.

Radiolabeling
The peptide conjugates used in the preclinical study were dissolved at 2 mg/mL in double-distilled H2O and were placed in 50 μL aliquots in separate Eppendorf Protein

Radiolabeling
The peptide conjugates used in the preclinical study were dissolved at 2 mg/mL in double-distilled H 2 O and were placed in 50 µL aliquots in separate Eppendorf Protein LoBind tubes which were subsequently stored at −20 • C. Labeling of peptide conjugates for preclinical tests was conducted as detailed below: Labeling with In-111. The following items were successively pipetted into an Eppendorf Protein LoBind ® centrifuge tube (capacity: 1.

Quality Control of Radiolabeled Products
For Preclinical Studies. Reverse-phase high-performance liquid chromatography (RP-HPLC) was the used for the quality control. A Waters Chromatograph based on a 600E multi-solvent delivery system and coupled to parallel photometric (Waters 2998 photodiode array detector; Vienna, Austria) and radiometric (Gabi gamma-detector from Raytest, RSM Analytische Instrumente GmbH; Straubenhardt, Germany) detection modes was used for analyses. The Empower Software (Waters, Milford, MA, USA) was used to control the system. For the analysis, aliquots of the radiolabeling solution were loaded on a Symmetry Shield RP18 cartridge column (5 µm, 3.9 mm × 20 mm, Waters, Eschborn, Germany), which was eluted with the following linear gradient: 100%A/0%B to 40%A/60%B in 20 min, whereby A = 0.01% TFA in H 2 O (v/v) and B = MeCN (system 1). The radiochemical purity of all radioligands was >98%. Hence, radioligand solutions were applied without further purification in biological assays that followed. The quality of radioligand solutions was tested before and after the conclusion of experiments.
Alternative methods for the preparation and the quality control of [ 177 Lu]Lu-AAZTA 5 -LM4, involving manual or module semi-automatic production modes as well as a radiochemical stability study, are included in the Supplementary Files. Moreover, descriptions of the preparation and quality control of [ 68 Ga]Ga-DATA 5m -LM4 and [ 177 Lu]Lu-AAZTA 5 -LM4 administered to the patient are also reported therein.

Preparation of [ nat In]In-AAZTA 5 -LM4 and [ nat Lu]Lu-AAZTA 5 -LM4
A stock solution (60 µL, 2 mM, 120 nmol) of the respective nitrate salt dissolved in 1 M sodium acetate buffer of pH 4.6 (for indium) and pH 5.0 (for lutetium) was added into an Eppendorf Protein LoBind ® centrifuge tube containing AAZTA 5 -LM4 precursor stock solution (60 µL, 120 µg, ≈60 nmol) and the mixture was heated at 75 • C for 1 h. Complete metal incorporation by AAZTA 5 -LM4 was revealed by HPLC adopting the PDA-UV detection mode. Analyte separations were achieved on a Waters XSelect CSH TM C18 reverse phase column (5 µm, 4.6 mm × 150 mm) applying the following linear gradient system at a 1.0 mL/min flow rate: starting from 10% B to 15% B in 5 min, followed by a further 0.5%/min increase in B within the ensuing 60 min (system 2; A and B, as given above in 2.1.3.). Application of these conditions allowed for baseline separation of metal-tagged from metal-free AAZTA 5 -LM4, as reported below. Retention times (UV trace, t R in min): AAZTA 5 -LM4, t R = at 20.6 min; [ nat In]In-AAZTA 5 -LM4, t R = 19.6 min; [ nat Lu]Lu-AAZTA 5 -LM4, t R = 19.4 min.
Only authorized personnel handled any solution containing beta-/gamma-emitting radionuclides in licensed facilities in compliance with European radiation safety guidelines. Protocols and facilities were supervised by the Greek Atomic Energy Commission (GAEC, license #A/435/17092/2019 and #A/435/15767/2019).

Cell Culture
Two cell lines were employed in the present study. Firstly, the wild-type (wt) HEK293 cells, lacking any measurable SST 2 R expression, served as negative controls. Then, HEK293 cells transfected to stably express the human SST 2 R tagged with the T7-epitope (HEK293-SST 2 R) were used, which were a kind gift of Prof. S. Schultz (Jena, Germany). Dulbecco's Modified Eagle Medium (DMEM), containing Glutamax-I and supplemented with 10% (v/v) heat-inactivated fetal bovine serum (FBS), 100 U/mL penicillin and 100 µg/mL streptomycin, was used as cell culture medium. Cells were kept in a controlled humidified atmosphere with 5% CO 2 at 37 • C. For transfected HEK293-SST 2 R cells, 400 µg/mL G418 was additionally introduced in the medium. Culture media were obtained from by Gibco BRL, Life Technologies (Grand Island, NY, USA) and supplements were purchased from Biochrom KG Seromed (Berlin, Germany). Cell passages were conducted at 70-85% confluency using a solution of trypsin/EDTA (0.05%/0.02% w/v).

Competition Binding Experiments
Competition binding assays were conducted for AAZTA 5 -LM4, [ nat In]In-AAZTA 5 -LM4 and [ nat Lu]Lu-AAZTA 5 -LM4 against [ 125 I-Tyr 25 ]LTT-SS28 in freshly harvested HEK293-SST 2 R cell membranes. In brief, to assay tubes test compound (30 µL solution of increasing 10 −13 -10 −5 M concentrations), radioligand (70 µL, 50 pM corresponding to ≈40,000 cpm) and membrane homogenate (200 µL) were added to a final volume of 300 µL in binding buffer (50 mM HEPES pH 7.4, 1% BSA, 5.5 mM MgCl 2 , 35 µM bacitracin). Samples in triplicate for each concentration point were incubated for 1 h at 22 • C in an Incubator-Orbital Shaker unit, (MPM Instr. SrI). Competition was abruptly stopped by the addition of ice-cold washing buffer and fast filtration through glass fiber filters (Whatman GF/B, pre-soaked for 2 h in a 1% polyethyleneimine (PEI) aqueous solution). This operation was contacted semi-automatically on a Brandel Cell Harvester (Adi Hassel Ingenieur Büro, Munich, Germany). They were subsequently rinsed with ice-cold washing buffer (10 mM HEPES pH 7.4, 150 mM NaCl). Filters were collected in plastic tubes and their radioactivity content was measured in an automated multi-sample well-type γ-counter (NaI(Tl) 3 crystal-equipped Canberra Packard Cobra TM Quantum U5003/1, Auto-Gamma ® instrument). Nonlinear regression for a one-site model was applied to determine the half-maximal inhibitory concentration (IC 50 ) values using the PRISM TM 6.0 GraphPad software (San Diego, CA, USA). Results from three independent experiments in triplicate were calculated and presented as mean IC 50 ± standard deviation (sd) values.

Radioligand Uptake in HEK293-SST 2 R Cells
The following radioligands were tested side by side: [ 111 In]In-AAZTA 5 -LM4 together with [ 111 In]In-DOTA-LM3 and [ 177 Lu]Lu-AAZTA 5 -LM4 together with [ 177 Lu]Lu-DOTA-LM3. A day before the experiment, HEK293-SST 2 R cells were measured and (1 × 10 6 /well) placed in poly-lysine-coated six-well plates in the incubator. The following day, the plates were placed on ice. They were then washed two times with ice-cold internalization medium (IM: DMEM Glutamax-I supplemented by 1% (v/v) FBS). The supernatant was removed by aspiration and the plates were left on the bench at ambient temperature. Fresh warmed medium was added (1.2 mL) to the plates, followed by a solution of test radioligand (50,000 cpm corresponding to 0.5 pmol total peptide in 150 µL 0.5% BSA PBS). Internalization medium was added in the upper three wells of each plate (150 µL, total) and a solution of TATE to a final concentration of 1 µM (150 µL, non-specific) in the lower three ones. The plates were left in the Incubator at 37 • C for 60 min and the incubation was interrupted by placing the plates on ice and collecting the medium. The cells were then washed with ice-cold 0.5% BSA-PBS (1 mL) and the washings collected. After incubation of cells (2 × 5 min) at ambient temperature in acid wash buffer (50 mM glycine in 0.1 M NaCl, pH 2.8), supernatants were collected (membrane-bound fraction). Cells were washed and aspirated with 0.5% BSA-PBS, then lysed by the addition of 1 N NaOH. Mechanical Pharmaceutics 2023, 15, 776 6 of 18 detachment of lysates allowed full collection in the corresponding tubes (internalized fraction). Samples were measured for their radioactivity content in the γ-counter. Cellassociated/internalized radioactivity was calculated vs. total added activity for each well. The respective specific values were determined by subtracting values in the presence of excess TATE (lower well triplicates-non-specific) from those without the addition of blocker (upper well triplicates-totals). Results from at least three independent experiments in triplicate were given as average ± sd values.

Stability Studies
For the assessment of metabolic stability of [ 111 In]In-AAZTA 5 -LM4 and [ 177 Lu]Lu-AAZTA 5 -LM4 in vivo, 6 healthy male Swiss albino mice were used (NCSR "Demokritos" Animal House, Athens, Greece; body weight: 30 ± 5 g). The animals received a 100 µL bolus via the tail vein containing either [ 111 In]In-AAZTA 5 -LM4 or [ 177 Lu]Lu-AAZTA 5 -LM4 (2.5-3 nmol of total conjugate in vehicle: saline/EtOH 9/1 v/v; these solutions corresponded to up to 11 MBq for In-111 and up to 74 MBq for Lu-177). Mice were euthanized 5 min post-injection (pi) and blood was rapidly retrieved from the heart with a pre-chilled 1 mL penicillin syringe. Blood samples were immediately placed in pre-chilled Eppendorf Protein LoBind ® tubes on ice, containing EDTA (40 µL, 50 mM Na 2 EDTA solution). Plasma was collected after centrifugation (10 min, 2000× g/4 • C, in a Hettich Universal 320R centrifuge, Tuttlingen, Germany), and an equal volume of cold MeCN was added. After a second centrifugation (10 min, 15,000× g/4 • C), supernatants were collected and concentrated to a small volume (≈50-100 µL) under a gentle N 2 flux at 40 • C. Physiological saline (400 µL) was added and the resulting solution samples were passed through a Millex GV filter (0.22 µm, 13 mm Ø, Millipore, Milford, MA, USA). Filtrate aliquots were analyzed by radio HPLC for the detection of forming radiometabolites (system 1). The elution time (t R ) of intact [ 111 In]In-AAZTA 5 -LM4 or [ 177 Lu]Lu-AAZTA 5 -LM4 was determined by co-injection of blood samples processed as above with an aliquot of the labeling solution. Results were obtained from three mice per radioligand and are presented as average percentage of intact radiopeptide ± sd.

Biodistribution in SCID Mice Bearing Twin HEK293-SST 2 R and wtHEK293 Tumors
Freshly harvested HEK293-SST 2 R cells (1.2 × 10 7 cells) and wtHEK293 cells (0.6 × 10 7 cells) were suspended in physiological saline (150 µL) and inoculated in the right and left flanks, respectively, of 43 male SCID mice (23.1 ± 1.6 g body weight, six weeks of age on arrival day; NCSR "Demokritos" Animal House, Athens, Greece). Mice were kept under aseptic conditions for 12-15 days and until palpable tumors (300-600 mg) were grown at the inoculation sites. On the day of biodistribution, animals were intravenously (iv) injected in groups of four with 100 µL bolus (vehicle: saline/EtOH 9/1 v/v) containing 74-92 kBq of either [ 111 In]In-AAZTA 5 -LM4 or [ 111 In]In-DOTA-LM3 and corresponding to 20-25 pmol AAZTA 5 -LM4 or DOTA-LM3, respectively. Animals were euthanized at 4 and 24 h pi and immediately dissected. Blood samples, organs of interest and tumors were rapidly collected, weighed and counted in the gamma counter. Likewise, a 100 µL bolus containing 730 kBq of either [ 177 Lu]Lu-AAZTA 5 -LM4 or [ 177 Lu]Lu-DOTA-LM3 and corresponding to 40 pmol AAZTA 5 -LM4 or DOTA-LM3 was iv injected in groups of four of inoculated mice. Animals were euthanized at 4, 24 and 48 h pi and the procedure described above was followed. Biodistribution data were calculated with the aid of suitable standards of the administered dose as percent of the injected activity per gram tissue (%IA/g) with excel software (Microsoft Corporation, Redmond, Washington, DC, USA). Results were provided as average ± sd values, n = 4 per time point, via the PRISM TM 6.0 GraphPad software (San Diego, CA, USA). Statistical analysis was performed as previously described [30].
Animal experiments were performed according to European and national regulations in certified facilities (EL 25 BIO exp021). The study protocols were approved by the Department of Agriculture and Veterinary Service of the Prefecture of Athens (approval of stability studies: #1609, 24 April 2019-approval of biodistribution and imaging studies #1610, 24 April 2019).

Patient Study
A 68-year-old female patient with a gastric NET of WHO grade III and a Ki67 index >15% with multiple lymph node and hepatic metastases was included in this study following her signed written informed consent.  [46], the patient underwent first PET/CT with [ 68 Ga]Ga-DATA 5m -LM4 on a dedicated GE Discovery 710 × 128 Slice PET/CT Scanner (GE HealthCare Technologies Inc.,/GE Healthcare, Chicago, IL, USA), with a 40 mm detector at a 0.35 s rotation speed and a 128-slice CT scanner. The images were corrected for random and scatter counts, decay correction and dead time correction. PET images were reconstructed with iterative reconstruction using ordered subset expectation maximization algorithm (OSEM) (2 iterations, 24 subsets). They were processed and analyzed with a dedicated commercially available workstation (GE Xeleris). Further details on the patient preparation, scan and data acquisition protocol are included in the Supplementary Materials.
Following injection of [ 177 Lu]Lu-AAZTA 5 -LM4, whole-body scans were performed on a Dual Head Gamma Camera (GE Discovery NM/CT 670; GE HealthCare Technologies Inc.,/GE Healthcare, Chicago, IL, USA) using a high-energy general purpose (HEGP) collimator. Scans were acquired in the Lu-177 window at a 20% energy window with a peak at 208 keV and 113 keV. Serial whole-body scans were acquired at 4, 17-24, 48, 96, 120 and 168 h pi. The matrix size for the whole-body scan was 256 × 1024. Anterior and posterior views were acquired by using both detectors. After acquiring the whole-body scan, serial SPECT/CT of the region of interest, i.e., abdomen and pelvis area in our study were acquired at 4, 17-24, 48, 96, 120 and 168 h pi. The scans were acquired in H mode, with a starting angle of 0 • . The matrix size used for the study was 128 × 128. The arc per detector was 180º and the number of views was 60. The scan mode was step and shoot, with a duration of 15 s/step. The CT scan involved a diagnostic dose CT with 300-380 mAs and 120 kVp, with a slice thickness of 2.5 mm and pitch of 0.6. The matrix size used for CT was 512 × 512. CT was used for attenuation correction as well as anatomical localization.

Ligands and Radioligands
Analytical data for the new AAZTA 5 -LM4 bioconjugate are presented in Tables S1 and S2 (Supplementary Materials). Table S1 contains data of HPLC analyses in two separate systems and MALDI-TOF MS data. Table S2 includes results from amino acid analysis. All data turned out to be consistent with the formation of AAZTA 5 -LM4 in high purity (≥99%). For the preclinical studies, both AAZTA 5 -LM4 and DOTA-LM3 were labeled with In-111 and Lu-177, as previously described [16,20,21,42]. Notably, radiolabeling of AAZTA 5 -LM4 with either radiometal proceeded by brief incubation at 50 • C (10 min for In-111 and 30 min for Lu-177). The final radiolabeled products were obtained in a ≥99% purity under these conditions at apparent molar activities of 3.7-7. Interestingly, alternative radiolabeling protocols for labeling of AAZTA 5 -LM4 with Lu-177 and comprising manual (at room temperature) and semi-automated module-involving procedures (at 50 • C) were successfully conducted as well (Supplementary Materials). Results from a kinetic study of [ 177 Lu]Lu-AAZTA 5 -LM4 formation at room temperature using different amounts of precursor (1-30 nmol) are summarized in Figure S2 Figure 3a for In-111 and 3b for Lu-177. The total specific cell uptake of [ 111 In]In-AAZTA 5 -LM4 (48.2 ± 1.8% of added activity) was found lower than that of [ 111 In]In-DOTA-LM3 (54.0 ± 3.6% of added activity; p < 0.0001). The same trend was observed for [ 177 Lu]Lu-AAZTA 5 -LM4 (47.7 ± 1.6% of added activity) and [ 177 Lu]Lu-DOTA-LM3 (61.9 ± 3.0% of added activity; p < 0.0001). Interestingly, the bulk of radioactivity for all tested analogs was found on the cell membrane, as for example in the case of [ 111 In]In-AAZTA 5 -LM4 (43.1 ± 1.7% membrane-bound fragment vs. 5.1 ± 0.3% internalized fragment). This pattern of cell distribution was consistent with SST 2 R-antagonist behavior. In all cases, cell uptake was markedly reduced in the presence of excess TATE, in accordance with an SST 2 R-mediated process.

Comparative Radioligand Uptake in HEK293-SST2R Cells
The

SPECT/CT of Mice Bearing Twin HEK293-SST 2 R and wtHEK293 Xenografts
SPECT/CT was performed for [ 111 In]In-AZTA 5 -LM4 in three SCID mice bearing double HEK293-SST 2 R and wtKEK293 tumors in their flanks at 4 (two animals) and 24 h pi (one mouse); images are depicted in Figure 5. [ 111 In]In-AZTA 5 -LM4 was selectively taken up only by the SST 2 R-expressing tumors, but not by the receptor-negative controls, verifying an SST 2 R-mediated process in agreement with biodistribution findings. Kidney uptake was intensive at 4 h pi, but significantly declined at 24 h pi, leading to favorable increase in Tu-to-Ki ratios concordant with biodistribution results.  Figure 5. [ 111 In]In-AZTA 5 -LM4 was selectively taken up only by the SST2R-expressing tumors, but not by the receptor-negative controls, verifying an SST2R-mediated process in agreement with biodistribution findings. Kidney uptake was intensive at 4 h pi, but significantly declined at 24 h pi, leading to favorable increase in Tu-to-Ki ratios concordant with biodistribution results.

Patient Study
First impressions from the tumor targeting and pharmacokinetic behavior of [ 177 Lu]Lu-AAZTA 5 -LM4 were acquired in a 68-year-old female patient with advanced stomach NET with extensive spread to the liver and lymph nodes. Representative imaging results are shown in Figure 6. orange arrows are pointing to HEK293-SST 2 R xenografts and light blue arrows to wtHEK293 tumors, devoid of SST 2 R expression. Intense uptake is observed in the SST 2 R-expressing tumors but no uptake is evident in the negative controls. Yellow arrows are directed toward the kidneys; the initial kidney uptake at (a) and (b) 4 h pi notably declines at (c) 24 h pi. The color bars indicate the difference in accumulated activity (purple being the lowest and white the highest level of accumulation).

Patient Study
First impressions from the tumor targeting and pharmacokinetic behavior of [ 177 Lu]Lu-AAZTA 5 -LM4 were acquired in a 68-year-old female patient with advanced stomach NET with extensive spread to the liver and lymph nodes. Representative imaging results are shown in Figure 6.  The patient initially underwent PET/CT with [ 68 Ga]Ga-DOTA-NOC [46], which revealed weak uptake in the liver metastases. PET/CT imaging with the SST2R-antagonist [ 68 Ga]Ga-DATA 5m -LM4 was much more successful in the visualization of hepatic metastases as early as 10 min pi and spanning up to 3 h pi. This clearly advantageous performance of the [ 68 Ga]Ga-DATA 5m -LM4 antagonist compared to the [ 68 Ga]Ga-DOTA-NOC agonist was the impetus to explore the overall biodistribution of the respective [ 177 Lu]Lu- The patient initially underwent PET/CT with [ 68 Ga]Ga-DOTA-NOC [46], which revealed weak uptake in the liver metastases. PET/CT imaging with the SST 2 R-antagonist [ 68 Ga]Ga-DATA 5m -LM4 was much more successful in the visualization of hepatic metastases as early as 10 min pi and spanning up to 3 h pi. This clearly advantageous performance of the [ 68 Ga]Ga-DATA 5m -LM4 antagonist compared to the [ 68 Ga]Ga-DOTA-NOC agonist was the impetus to explore the overall biodistribution of the respective [ 177 Lu]Lu-AAZTA 5 -LM4 in this patient applying sequential SPECT/CT imaging in a period of 4 h-72 h pi. At this early stage of the study, the priority was set to explore the uptake and retention of the new antagonist in the lesions, a critical issue for its value and applicability for therapy. In parallel with the exceptional results of [ 68 Ga]Ga-DATA 5m -LM4 PET/CT, the new radioligand [ 177 Lu]Lu-AAZTA 5 -LM4 displayed high tumor uptake, which peaked at 17 h pi. Some uptake was observed in respiratory mucosa, spleen and kidneys, with rather low uptake in the liver. Radioactivity was excreted predominantly via the urinary system, with some portion of hepatobiliary excretion being observed. Accordingly, the highest tumor-to-background ratio was achieved at 24 h pi. These preliminary clinical results need to be confirmed by more patient cases, and careful dosimetric studies must be conducted prior to establishing the clinical value of [ 177 Lu]Lu-AAZTA 5 -LM4 as a useful radionuclide therapy candidate in NET patients.

Discussion
We recently reported on [ 68 Ga]Ga-DATA 5m -LM4, a new SST 2 R-antagonist radiotracer for application in the diagnosis of human NETs with PET/CT [30]. Coupling of the hybrid chelator DATA 5m at the N-terminus of LM4 allowed for fast incorporation of radiogallium (Ga-67 or Ga-68) at conveniently lower temperatures than those required for DOTA-modified vectors, such as DOTA-LM3. Thus, [ 67 Ga]Ga-DATA 5m -LM4 was easily accessible and showed exceptional preclinical features in HEK293-SST 2 R/wtHEK293 cells and tumorbearing mice when compared with [ 67 Ga]Ga-DOTA-LM3. Moreover, in a first proof-ofprinciple study in a NET patient, [ 68 Ga]Ga-DATA 5m -LM4 was able to visualize tumor lesions accurately and with a high contrast on PET/CT. We have now attached the AAZTA 5chelator in place of DATA 5m to LM4 to allow for facile labeling with a broader palette of trivalent metals of medical interest [34]. According to recent reports, AAZTA 5 -derivatized vectors could be successfully labeled with Lu-177, Sc-44 or In-111 at lower temperatures. The subsequent study of resulting analogs has brought to light the easiness of AAZTA 5 methodology to provide ready-to-use radioligands in a clinical setting [33,35,[38][39][40][41].
In the present work, AAZTA 5 -LM4 and DOTA-LM3 were labeled with In-111 and Lu-177 and the biological profiles or forming radiopeptides were directly compared in HEK293-SST 2 R/wtHEK293 cells and tumors thereof raised in mice. Typically, higher and longer heating was required for full incorporation of In-111 and Lu-177 by DOTA-LM3 (80 • C for 30 min), as opposed to AAZTA 5 -LM4, which could be labeled at less drastic conditions (50 • C for 10 min for In-111, 30 min for Lu-177). It should be noted that equally successful labeling of AAZTA 5 -LM4 with Lu-177 could be achieved manually at room temperature or via a module semi-automatic process preset at 50 • C. Exceptional reproducibility was confirmed across these methods, whereby [ 177 Lu]Lu-AAZTA 5 -LM4 could be easily obtained in >98% radiochemical purity at a molar activity of 40 MBq/nmol.
It is interesting to observe that the SST 2 R binding affinity of AAZTA 5 -LM4 (IC 50 = 1.69 ± 0.47 nM) slightly improved after coordination of indium ([ nat In]In-AAZTA 5 -LM4 IC 50 = 0.45 ± 0.05 nM) or lutetium ([ nat Lu]Lu-AAZTA 5 -LM4 IC 50 = 0.55 ± 0.38 nM). These values fall well within the range previously determined by the same assay for DATA 5m -LM4 (IC 50 = 1.24 ± 0.20 nM) and [ nat Ga]Ga-DATA 5m -LM4 (IC 50 = 1.61 ± 0.32 nM) [30]. It should be noted that previous studies with DOTA-LM3 (IC 50 = 1.4 ± 0.5 nM) revealed striking changes in SST 2 R affinities upon coordination of gallium (IC 50 = 12.5 ± 0.4.3 nM), but not lutetium (IC 50 = 1.6 ± 0.3 nM); further changes were observed as well by switching the chelator from DOTA to NODAGA in LM3 [16,47,48]. Thus, a strong impact of metal, chelator or metal-chelate on SST 2 R affinity was evident in the case of LM3, with such differences being practically absent in DATA 5m /AAZTA 5 Figure 3). Interestingly, [ 67 Ga]Ga-DATA 5m -LM4 had previously demonstrated much superior uptake compared with [ 67 Ga]Ga-DOTA-LM3, which was poorly taken up in the same cells [30]. This discrepancy is consonant with SST 2 R affinity differences and further underscores the impact of metal-chelate on the biological responses of LM3 radioligands. It should be stressed that in all cases, the bulk of radioactivity was found attached to the cell membrane, with only a small fraction being internalized, consonant with an SST 2 R-antagonist profile.
After injection in animals, both [ 111 In]In-AAZTA 5 -LM4 and [ 177 Lu]Lu-AAZTA 5 -LM4 were found to be metabolically robust, similarly to [ 67 Ga]Ga-DATA 5m -LM4 [30]. It appears that LM4-based radioligands withstand the degrading action of major proteolytic enzymes, such as neutral endopeptidase or angiotensin-converting enzyme, implicated in the rapid catabolism of many radiopeptides in vivo and compromising their delivery to tumor sites [49].  [30], we observe that the respective two In-111 radiotracers tested herein displayed higher tumor uptake and much lower renal retention. Interestingly, the most striking differences were found between [ 111 In]In-DOTA-LM3 and [ 67 Ga]Ga-DOTA-LM3 for the tumor (In: 36.38 ± 2.05%IA/g and Ga: 16.83 ± 1.22%IA/g) and kidneys (In: 15.36 ± 0.58%IA/g and Ga: 37.65 ± 3.44%IA/g) at 4 h pi. These results favor a future application of [ 111 In]In-AAZTA 5 -LM4, and possibly also [ 111 In]In-DOTA-LM3, as alternative diagnostic tracers for SPECT/CT, by allowing for imaging at later time points associated with higher tumor-to-background ratios. Furthermore, they offer the option of SST 2 R-targeted diagnosis of NETs in nuclear medicine facilities lacking PET/CT instrumentation.
Likewise, the biodistribution of [ 177 Lu]Lu-AAZTA 5 -LM4 and [ 177 Lu]Lu-DOTA-LM3 were directly compared in the same twin HEK293-SST 2 R/wtHEK293 tumor-bearing animal model (Figure 4b). In line with in vitro cell uptake findings, [ 177 Lu]Lu-DOTA-LM3 displayed higher uptake in the HEK293-SST 2 R xenografts compared with [ 177 Lu]Lu-AAZTA 5 -LM4 at all time points tested. However, the background radioactivity levels were unfavorably much higher for [ 177 Lu]Lu-DOTA-LM3 in almost all organs and tissues, except for the kidneys. It is not easy to predict to what extent this pattern could be attributed to affinity differences between the mice and human receptors, or if it can be altered by the administration of higher peptide amounts or kidney protection regimens, or if it can be extrapolated from mice to humans. Further dedicated studies are warranted to properly address these questions, which are essential for dosimetry calculations and rational therapy planning.
In a first step toward this goal, we then investigated the biodistribution of [ 177 Lu]Lu-AAZTA 5 -LM4 in a patient with a previously confirmed gastric NET metastasized in the liver and lymph nodes. Initial PET/CT with the established SST 2 R-agonist [ 68 Ga]Ga-DOTA-NOC [46] failed to accurately reveal most liver lesions as opposed to PET/CT with the SST 2 R-antagonist [ 68 Ga]Ga-DATA 5m -LM4. The latter successfully visualized hepatic lesions with high contrast, confirming previous reports on the superior diagnostic power of SST 2 R-antagonists compared with agonists [10,[12][13][14][15]. Based on this promising outcome, the patient was next injected with [ 177 Lu]Lu-AAZTA 5 -LM4 and followed by SPECT/CT for tumor uptake and retention and overall tissue distribution pattern up to 72 h pi. The radioligand was well tolerated by the patient, showing high and prolonged uptake in tumor lesions. In contrast, the radioactivity declined faster from physiological tissues, leading to favorable tumor-to-background ratios at 24 h pi. Excretion was mainly achieved via the urinary system, with the kidneys still visible at 72 h pi. The overall pattern of [ 177 Lu]Lu-AAZTA 5 -LM4 in the patient turned out to be quite promising for radionuclide therapy, after further study of human dosimetry, safety, therapeutic efficacy and selection of a suitable therapeutic scheme. Ongoing studies will eventually establish its actual therapeutic value in the clinic.

Conclusions
Derivatization of the SST 2 R-antagonist LM4 with the hybrid chelator AAZTA 5 has allowed convenient labeling with In-111 (for diagnostic SPECT/CT) and Lu-177 (for radionuclide therapy), expanding the application prospects of LM4 beyond [ 68 Ga]Ga-DATA 5m -LM4 PET/CT, previously proposed. The hybrid character of DATA 5m /AAZTA 5