Synthesis and Evaluation of Two Long-Acting SSTR2 Antagonists for Radionuclide Therapy of Neuroendocrine Tumors

Somatostatin receptor subtype 2 (SSTR2) has become an essential target for radionuclide therapy of neuroendocrine tumors (NETs). JR11 was introduced as a promising antagonist peptide to target SSTR2. However, due to its rapid blood clearance, a better pharmacokinetic profile is necessary for more effective treatment. Therefore, two JR11 analogs (8a and 8b), each carrying an albumin binding domain, were designed to prolong the blood residence time of JR11. Both compounds were labeled with lutetium-177 and evaluated via in vitro assays, followed by in vivo SPECT/CT imaging and ex vivo biodistribution studies. [177Lu]Lu-8a and [177Lu]Lu-8b were obtained with high radiochemical purity (>97%) and demonstrated excellent stability in PBS and mouse serum (>95%). [177Lu]Lu-8a showed better affinity towards human albumin compared to [177Lu]Lu-8b. Further, 8a and 8b exhibited binding affinities 30- and 48-fold lower, respectively, than that of the parent peptide JR11, along with high cell uptake and low internalization rate. SPECT/CT imaging verified high tumor accumulation for [177Lu]Lu-8a and [177Lu]Lu-JR11 at 4, 24, 48, and 72 h post-injection, but no tumor uptake was observed for [177Lu]Lu-8b. Ex vivo biodistribution studies revealed high and increasing tumor uptake for [177Lu]Lu-8a. However, its extended blood circulation led to an unfavorable biodistribution profile for radionuclide therapy.


Introduction
Treatment of neuroendocrine tumors (NETs) largely depends on radioligands targeting somatostatin receptor type 2 (SSTR2). [ 177 Lu]Lu-DOTA-TATE (Lutathera ® ) is the leading radioligand with approval from both the Food and Drug Administration (FDA) and the European Medicines Agency (EMA) [1]. The NETTER-1 phase III study showed promising results, with a response rate of 18% for the [ 177 Lu]Lu-DOTA-TATE group [2]. However, novel developments are still necessary to achieve a better response. One such strategy is to enhance the radioligand delivery to increase the radiation dose to tumor cells.
Various studies have shown that SSTR2 antagonists, such as JR11, are more potent than SSTR2 agonists due to their ability to bind to more receptor binding sites. Therefore, several SSTR2 antagonist peptides have been labeled for diagnostic or therapeutic purposes [3][4][5][6][7]. However, the rapid blood clearance of such SSTR2 radioligands and the significant accumulation in non-target tissues pose a limit for higher tumor dose delivery and more efficient treatment [8]. In this context, binding of radioligands to serum protein can be an efficient method to improve the pharmacokinetic properties of these molecules [9].
Albumin binding domains (ABD) promise to increase the time-integral tumor uptake by prolonging blood circulation and reducing the uptake in healthy organs such as the kidneys [10]. Recently, Evans Blue (EB), a molecule known to bind to albumin, was conjugated to the agonist octreotate, and the resulting EB-TATE was labeled with the therapeutic radionuclide yttrium-90. [ 90 Y]Y-EB-TATE showed higher tumor uptake and improved tumor response in mice bearing SSTR2-positive tumors compared to [ 177 Lu]Lu-DOTA-TATE [11]. Other ABDs with different binding affinity for albumin have also been reported [12][13][14]. Of those, 4-(p-iodophenyl)butyryl and 4-(p-methoxyphenyl)butyryl were the most preferred albumin binding domains due to the aforementioned studies, in which enhanced tumor uptake and tumor-to-kidney dose ratio were observed. Thus, we report herein the synthesis of two JR11 derivatives containing one of these ABDs on the side chain of a lysine residue incorporated into the peptide sequence. Then, the in vitro characteristics of these longacting SSTR2-antagonists (8a and 8b) were evaluated. Both compounds were labeled with lutetium-177, and their in vivo distribution in tumor xenografts, overexpressing SSTR2 were investigated by SPECT/CT imaging and ex vivo biodistribution.

Synthesis of the Long-Acting SSTR2 Antagonists
The synthesis of the two long-acting JR11 analogs 8a and 8b was carried out following standard Fmoc-based SPPS protocols (Scheme 1). Coupling of the chelator was performed using PyBOP, as we previously observed faster reaction kinetics when using this coupling agent [15,16]. Cleavage of the peptides from the solid support resulted in the removal of most of the side chain protecting groups. However, a second treatment with neat TFA was required for the complete deprotection of the tert-butyl groups. Further, ivDde protection of the sidechain amino group of the lysine residue in position-3 allowed orthogonal coupling of the albumin binding domains on the additional lysine positioned between the cyclic peptide and the chelator. Conjugation of the ABDs to 6 was performed after their activation as NHS esters. The final products 8a and 8b were obtained at 9.9 and 8.7% yield, respectively, after deprotection of the ivDde protecting group of Lys 3 and purification by HPLC ( Figures S1 and S2).

Radiolabeling with [ 177 Lu]LuCl 3
Labeling of 8a and 8b with [ 177 Lu]LuCl 3 was performed in the presence of gentisic acid and ascorbic acid. Both additives are known scavengers to protect radiopharmaceuticals from radiolytic degradation [17]. Kolliphor, a non-ionic solubilizer and emulsifier used to improve the solubility of hydrophobic compounds, was added to the labeling reaction mixture to prevent stickiness of the peptides. The radiochemical yield (RCY) of both peptides and their radiochemical purity (RCP) are presented in Table 1

Radiolabeling with [ 177 Lu]LuCl3
Labeling of 8a and 8b with [ 177 Lu]LuCl3 was performed in the presence of gentisic acid and ascorbic acid. Both additives are known scavengers to protect radiopharmaceuticals from radiolytic degradation [17]. Kolliphor, a non-ionic solubilizer and emulsifier used to improve the solubility of hydrophobic compounds, was added to the labeling reaction mixture to prevent stickiness of the peptides. The radiochemical yield (RCY) of both peptides and their radiochemical purity (RCP) are presented in Table 1

Competitive Binding Assay
The IC 50 values of 8a and 8b were determined in a competitive binding assay using U2OS-SSTR2 cells and [ 177 Lu]Lu-JR11 as radioligand (Table 2 and Figure S6); 8a and 8b exhibited a binding affinity of 80 and 130 nM, respectively, which is 30-and 48-fold lower than the IC 50 value found for the parent peptide JR11. Considering that the binding affinities of 8a and 8b were still in the nanomolar range, we considered that they were suitable for further studies. Uptake of 177 Lu-labeled 8a and 8b was observed in an in vitro assay using U2OS-SSTR2 cells (7.8 ± 0.05 and 3.1 ± 0.33% added dose, respectively ( Figure 1A). Most radioactivity uptake was membrane-bound in both cases (5.8 ± 0.17 and 2.6 ± 0.38% added dose, respectively), confirming the antagonist properties of the two compounds. However, compared to the [ 177 Lu]Lu-JR11 reference (total uptake 16.2 ± 3.0% added dose), this uptake was significantly lower, especially for [ 177 Lu]Lu-8b. A similar radioactive uptake pattern (930 ± 37 DLU/mm 2 for [ 177 Lu]Lu-JR11, 561 ± 18 DLU/mm 2 for [ 177 Lu]Lu-8a, and 280 ± 23 DLU/mm 2 for [ 177 Lu]Lu-8b) was observed when H69 human carcinoma tissues were incubated with the aforementioned compounds, as determined by autoradiography ( Figure 1B).

Ex Vivo Biodistribution Analysis
Next to the in vivo SPECT/CT imaging studies, ex vivo biodistribution was carried out after administration of [ 177 Lu]Lu-JR11 and [ 177 Lu]Lu-8a at the same time points (4, 24, 48, and 72 h). Compound 8b was excluded due to the minimal tumor uptake observed earlier in the in vivo SPECT scans. An additional group of mice per compound received an excess of the corresponding non-radioactive compound (block group) to determine receptor specificity. At 4 h post-injection, the uptake of [ 177 Lu]Lu-8a in the blood was at 21.3 ± 1.1% ID/g and gradually decreased to 8.6 ± 0.5% ID/g at 72 h post-injection, reflecting the albumin-binding properties of the compound ( Figure 3A and Table S1). Tumor uptake of [ 177 Lu]Lu-8a slightly increased over time (4.1 ± 0.5% ID/g at 4 h, 6.5 ± 0.2% ID/g at 24 h, 7.3 ± 0.4% ID/g at 48 h, and 7.3 ± 0.9% ID/g at 72 h). However, the kidneys showed high  Table 3. SPECT quantification of xenograft tumor (T) and kidneys (K) per group at 4, 24, 48, and 72 h post-injection, represented in % ID/mL.

Ex Vivo Biodistribution Analysis
Next to the in vivo SPECT/CT imaging studies, ex vivo biodistribution was carried out after administration of [ 177 Lu]Lu-JR11 and [ 177 Lu]Lu-8a at the same time points (4, 24, 48, and 72 h). Compound 8b was excluded due to the minimal tumor uptake observed earlier in the in vivo SPECT scans. An additional group of mice per compound received an excess of the corresponding non-radioactive compound (block group) to determine receptor specificity. At 4 h post-injection, the uptake of [ 177 Lu]Lu-8a in the blood was at 21.3 ± 1.1% ID/g and gradually decreased to 8.6 ± 0.5% ID/g at 72 h post-injection, reflecting the albumin-binding properties of the compound ( Figure 3A and Table S1). Tumor uptake of [ 177 Lu]Lu-8a slightly increased over time (4.1 ± 0.5% ID/g at 4 h, 6.5 ± 0.2% ID/g at 24 h, 7.3 ± 0.4% ID/g at 48 h, and 7.3 ± 0.9% ID/g at 72 h). However, the kidneys showed high uptake at the early and late time points (9.6 ± 1.2% ID/g at 4 h and 21.6 ± 1.4% ID/g at 72 h), suggesting clearance of the compound through renal excretion. Shortly after injection, a substantial level of radioactivity was also found in the lungs, pancreas, skin, and spleen. Finally, only a slight reduction was observed in the tumor after injection of the blocking agent.
Pharmaceuticals 2022, 15, x FOR PEER REVIEW 7 of 17 uptake at the early and late time points (9.6 ± 1.2% ID/g at 4 h and 21.6 ± 1.4% ID/g at 72 h), suggesting clearance of the compound through renal excretion. Shortly after injection, a substantial level of radioactivity was also found in the lungs, pancreas, skin, and spleen. Finally, only a slight reduction was observed in the tumor after injection of the blocking agent. In contrast, the reference compound [ 177 Lu]Lu-JR11 showed high tumor uptake at 4 h post-injection (8.4 ± 0.5% ID/g, Figure 3B and Table S2) and even though the tumor uptake slightly decreased over time (6.1 ± 0.5% ID/g at 24 h, 4.6 ± 0.3% ID/g at 48 h, and 3.6 ± 0.4% ID/g at 72 h post-injection), the tumor-to-kidney ratio increased from 0.6 ± 0.05 at 4 h post-injection to 1.2 ± 0.3 at 72 h post-injection. Complete blocking of tumor uptake (Block group) at 24 h post-injection confirmed the specificity of the JR11 compound for the SSTR2 receptor. In contrast, the reference compound [ 177 Lu]Lu-JR11 showed high tumor uptake at 4 h post-injection (8.4 ± 0.5% ID/g, Figure 3B and Table S2) and even though the tumor uptake slightly decreased over time (6.1 ± 0.5% ID/g at 24 h, 4.6 ± 0.3% ID/g at 48 h, and 3.6 ± 0.4% ID/g at 72 h post-injection), the tumor-to-kidney ratio increased from 0.6 ± 0.05 at 4 h post-injection to 1.2 ± 0.3 at 72 h post-injection. Complete blocking of tumor uptake (Block group) at 24 h post-injection confirmed the specificity of the JR11 compound for the SSTR2 receptor.

Discussion
Somatostatin receptor subtype 2 (SSTR2) is present at a high incidence in neuroendocrine tumors (NETs) and therefore is an ideal target for imaging and therapy of this malignant disease. Somatostatin analogs have been widely used for the past decades and are considered a gold standard for NET treatment [19]. Among these drugs, radiolabeled somatostatin antagonists such as LM3, JR10, and JR11 have shown greater promise than agonists (e.g., DOTA-TATE and DOTA-TOC) [20]. However, preclinical and clinical studies have demonstrated that JR11 is cleared rapidly from the bloodstream, thus leading to high kidney uptake [7,21]. Therefore, our study aimed to develop two JR11 analogs carrying two different albumin binding domains to improve the pharmacokinetic profile of JR11.
The chemical structure of our new ligands (8a and 8b) is directly based on the structure of the parent peptide JR11. The introduction of the albumin binding domains into the peptide sequence was established by the attachment of a lysine residue between the cyclic peptide and the DOTA chelator. This method has been previously employed to introduce fluorescent dye on the chemical structure of existing radioligands [22][23][24]. Here, we investigated the influence of two ABDs, namely 4-(p-iodophenyl)butyryl and 4-(pmethoxyphenyl)butyryl, on the in vivo behavior of JR11. These ABDs were selected due to the promising results reported in previous studies [12][13][14]25]. The 4-(p-iodophenyl)butyryl group has been widely used to improve the bioavailability of different radioligands, such as DOTA-TATE, PSMA-617, and folic acid. However, when conjugated to PSMA-617 ([ 177 Lu]Lu-HTK01169), Hsiou-Ting Kuo et al. noticed that not only was tumor uptake 8.3-fold higher for [ 177 Lu]Lu-HTK01169 in comparison to [ 177 Lu]Lu-PSMA-617, but also, the absorbed dose in the kidneys was 17.1-fold higher than that of the parent molecule [26]. Later, the same group reported a study comparing several albumin binding domains and concluded that 4-(p-methoxyphenyl)butyryl could be a potential candidate to improve blood circulation and reduce kidney radiotoxicity [13]. Radiolabeling of 8a and 8b with [ 177 Lu]LuCl 3 was successful, and both radiopeptides showed high RCYs and RCPs. They exhibited excellent inertness in PBS and mouse serum up to 24 h post-incubation, proving their stability towards radiolysis and peptidase digestion. When compared to JR11, the LogD 7.4 values of our radiopeptides were slightly higher than the LogD 7.4 value of the parent peptide [18]. This is probably due to the lipophilic character of the two albumin binding domains. This observation is also confirmed by the data provided by Hsiou-Ting Kuo et al. [13]. Nevertheless, the LogD 7.4 values of [ 177 Lu]Lu-8a and [ 177 Lu]Lu-8b remained negative, suggesting that they are hydrophilic. [ 177 Lu]Lu-8a showed good binding to plasma proteins, as previously reported for other radioligands bearing the same ABD [27,28]. Furthermore, [ 177 Lu]Lu-8a showed stronger affinity towards human albumin in comparison to that of [ 177 Lu]Lu-8b, confirming that the interaction between the ABD and the plasma proteins is influenced by the lipophilicity of the substituted phenyl group [13]. Thus, two JR11 analogs were prepared and, as expected, 8a interacted strongly with albumin, while 8b exhibited a milder interaction with albumin. The parent peptide JR11 showed lower binding to plasma proteins in comparison to that of our long-acting analogs.
Competitive binding assays were performed to determine the IC 50 values of the newly synthesized compounds 8a and 8b for the SSTR2 receptor. The incorporation of the albumin binding moiety did affect the SSTR2 binding affinity of the two compounds, probably due to the sensitivity of the parent peptide JR11 to chemical modifications. Fani et al. noticed that different chelators affect the binding affinity of the peptide in vitro [7]. More specifically, exchanging the DOTA chelator with NODAGA ([ 68 Ga]Ga-NODAGA-JR11) significantly increased the affinity of the peptide toward the SSTR2 receptor. The authors speculated that in the case of the DOTA chelator, the geometry is hexacoordinate, and in solution, the chelator can act as a spacer, hence lowering the affinity. In vitro uptake of these compounds in cells and tumor sections expressing SSTR2 was lower than the uptake of JR11. The majority of the uptake was found in the membrane-bound fraction, confirming the compounds' antagonist properties.
SPECT/CT images revealed increased blood residence time of the compound [ 177 Lu]Lu-8a compared with [ 177 Lu]Lu-JR11. This suggests that the albumin binding affinity contributed to the different pharmacokinetic profiles. However, despite the higher tumor uptake, as observed by image analysis, accumulation in the kidneys was substantial, thus making it unsuitable for future therapeutic studies. In a different study, Rousseau et al. also observed the same pattern when comparing DOTA-TATE with an albumin binding moiety ([ 177 Lu]Lu-AspAB-DOTA-TATE) to the reference analog, highlighting that albumin binding moieties might negatively affect the pharmacokinetic profile of peptides [25]. In the clinic, the high kidney uptake can be reduced by perfusion of cationic amino acids, but bone marrow toxicity will remain a problem [29]. Unfortunately, and to our surprise, compound [ 177 Lu]Lu-8b showed no tumor uptake and rapid renal clearance despite earlier reports by Kuo et al. showing the opposite effect with PSMA ligands [13].
Ex vivo biodistribution analysis confirmed the uptake pattern observed during imaging. In addition, the introduction of a blocking group for both 8a and JR11 showed non-specific tumor uptake for [ 177 Lu]Lu-8a compared to [ 177 Lu]Lu-JR11, for which a total blockade of SSTR2 receptor uptake was achieved. The high uptake of the compound in most organs could suggest that the plasma protein-binding prolongs the blood circulation substantially and interferes with the binding of 8a to tumor cells, thus causing inefficient blockade. Müller and coworkers also used the 4-(p-iodophenyl)butyryl ABD conjugated to a folate radioligand, and they observed higher tumor uptake and a reduction in kidney accumulation [14]. However, the authors did not perform a blocking study, making it difficult to draw any conclusions on the specificity of their compound. On the other hand, van Tiel et al. did not observe total blockage of tumor uptake when they conjugated the same albumin binding domain onto Albutate-1 [12]. More specifically, the tumor uptake of [ 177 Lu]Lu-Albutate-1 was 24.42 ± 1.44% ID/g at 24 h post-injection, while the blocked group showed uptake of approximately 12% ID/g. From dosimetry calculations, the authors noticed a high radiation dose in several tissues, especially in bone marrow (total absorbed dose of 765 mGy/MBq), probably due to the prolonged circulation. Even though in this study bone marrow uptake was negligible due to the inadequate extraction of the sample during the ex vivo biodistribution, this could pose a limit for a better therapeutic index with compounds carrying albumin binding moieties.
In a more recent study, the 4-(p-iodophenyl)butyryl moiety was conjugated to the DOTA-(PEG 28 ) 2 -A20FMDV2 peptide labeled with [ 177 Lu]Lu and used for the imaging of αvβ6-positive tumors [30]. The authors noticed that even though blood uptake of the peptide was high at 1 h post-injection, it dropped rapidly by 48 h (0.04 ± 0.01% ID/g), while tumor uptake remained constant (4.06 ± 0.54% ID/g), rendering the tumor-to-blood ratio ideal for therapy. Kidney uptake was, however, responsible for the toxicity observed during these studies. Based on this, it is recommended to perform the blocking study at a later timepoint to allow for better clearance of the peptide from the blood.
Overall, the previously mentioned studies, together with our results, point out the need for different and better albumin binding moieties and further modifications to the peptide itself that will improve the binding affinity and the specificity of SSTR2-targeting antagonists towards the receptor.

Determination of Lipophilicity
The distribution coefficient (LogD 7.4 ) of the 177 Lu-labeled peptides was determined by the shake-flask method. For each radioligand, the experiment was performed in triplicate. The radiolabeled compound (~1.5 MBq) was added to a solution of phosphate buffered saline (PBS)/n-octanol (1 mL, v:v = 1:1) in eppendorf vials. The vials were vortexed vigorously and centrifuged at 10,000 rpm for 3 min. The n-octanol phase was separated from the aqueous phase, poured into new vials, and centrifuged for 15 min. Samples from each phase (10 µL) were measured using a gamma counter. The LogD 7.4 value was calculated using the following equation: LogD 7.4 = ([counts in the n-octanol phase]/[counts in the PBS phase]).

Stability Studies in PBS and Mouse Serum
The 177 Lu-labeled compounds (~3 MBq) were incubated in PBS (300 µL) at 37 • C. The stability of the radiolabeled peptides was verified by radio-HPLC at 1, 4, and 24 h. The stability in serum was determined by incubating the radiolabeled compounds (~3 MBq) into 150 µL of mouse serum (Merck; Haarlerbergweg, The Netherlands) at 37 • C. At 1, 4, and 24 h post-incubation, the proteins were precipitated by adding an aliquot of 35 µL of the mixture to an equal volume of ACN. The vial was vortexed and centrifuged for 20 min. stability was monitored by radio-HPLC ( Figures S4 and S5).

Albumin Binding Properties
Compounds 8a, 8b, and JR11 were radiolabeled with lutetium-177 at a molar activity of 50 MBq/nmol. Radiopeptides (~1 MBq) were incubated in either PBS (500 µL) or human albumin/PBS (500 µL, v:v = 1:4) for 1 h at 37 • C. Three aliquots of 10 µL were counted in a gamma counter to determine the amount of activity in the loading solution. The mixtures were loaded onto Centrifree Ultrafiltration devices (Centrifree Ultrafiltration device with Ultracel PL membrane, 30 KDa, Merck, Haarlerbergweg, The Netherlands) preconditioned with 700 µL of kolliphor in PBS (0.06 mg/mL). The Centrifree Ultrafiltration devices were centrifuged at 7000 rpm for 30 min. Three aliquots of 10 µL from each mixture were counted in a gamma counter, and the protein-bound fraction was calculated based on the radioactivity measured in the filtrate relative to the corresponding loading solution.

Competitive Binding Assay
Competitive binding experiments against [ 177 Lu]Lu-JR11 were performed with 8a and 8b in U2OS-SSTR2 cells. Cells were seeded in a 24-well plate 24 h in advance (2 × 10 5 cells/well). On the day of the experiment, medium was removed, and the cells were washed once with PBS (Gibco). Then, solutions containing unlabeled compound 8a or 8b in increasing concentrations (10 −12 to 10 −5 M) in internalization medium (DMEM media, 20 mM HEPES, 1% BSA, pH 7.4) were added, followed by the addition of [ 177 Lu]Lu-JR11 (10 −9 M). For each concentration, experiments were performed in triplicate. Cells were incubated at 37 • C for 90 min. After incubation, medium was removed, and cells were washed once with PBS and lysed with 0.5 M sodium hydroxide (NaOH) for 10 min at rt. The lysate was transferred to counting tubes, and measurement was performed using the γ-counter.

Uptake and Internalization Assay
Cells were seeded in 6-well plates 48 h before the experiment (2 × 10 5 cells/well). The following day, adhered cells were incubated with 10 −9 M of [ 177 Lu]Lu-JR11, [ 177 Lu]Lu-8a, or [ 177 Lu]Lu-8b in 1 mL of culture medium for 2 h at 37 • C. After incubation, medium was removed, and cells were washed twice with PBS. The membrane-bound fraction was collected by incubating cells with an acid buffer (50 mM glycine, 100 mM sodium chloride (NaCl), pH 2.8) for 10 min at rt. Cells were lysed using 0.5 M NaOH for 10 min at rt to acquire the internalized fraction. Both fractions were counted in a γ-counter, and data were expressed as percentage of added dose.

Uptake and Autoradiography of H69 Tumor Sections
Subcutaneous fresh frozen H69 tumor tissues were cut at 10 µm thickness and immediately mounted on Starfrost glass slides (Thermo Scientific; Bleiswijk, The Netherlands). Tissue sections were incubated with washing buffer (167 mM Tris-HCl, 5 mM MgCl 2 ) containing 0.25% BSA for 10 min at rt to prevent nonspecific binding. Then, each section was incubated with 10 −9 M of [ 177 Lu]Lu-JR11, [ 177 Lu]Lu-8a or [ 177 Lu]Lu-8b diluted in incubation buffer (washing buffer containing 1% BSA) for 90 min at rt. Each slide was drained off and washed with PBS. Finally, the slides were exposed to super-resolution phosphor screens for 48 h and imaged with the Cyclone system (PerkinElmer; Waltham, MA, USA). Images were analyzed and quantified using the Optiquant software Version 5 (PerkinElmer; Waltham, MA, USA).

In Vivo Studies
All animal experiments were approved by the Animal Welfare Committee of the Erasmus MC, and all procedures were conducted according to accepted guidelines. Mice were subcutaneously inoculated with 5 × 10 6 SSTR2-positive H69 human small cell lung carcinoma cells in Matrigel, and tumors were left to grow for 3-4 weeks to an average volume of approximately 300 mm 3

SPECT/CT Imaging
The small-animal VECTor 5 /CT (MILAbs B. V.; Utrecht, The Netherlands) was used for all imaging studies. Image acquisition was performed at 4, 24, 48, and 72 h post-injection using the high-energy general-purpose mouse collimator (HE-GP-M, 0.8 mm pinhole size) in list mode. Corresponding CT scans were acquired in total-body and normal mode (50 kV, 0.21 mA, 75 ms) for anatomical reference and attenuation correction. All SPECT images were reconstructed using the similarity-regulated ordered-subsets expectation maximization (SROSEM) algorithm (MILAbs Rec 11.00 software, MILAbs, Utrecht, The Netherlands) with 5 iterations, 128 subsets, and a voxel size of 0.4 mm 3 . Image processing and analyses of the reconstructed data were performed using the PMOD image analysis software version 3.10 (PMOD Technologies; Zurich, Switzerland) to calculate the percentage of injected dose per mL (% ID/mL). To allow quantification of the SPECT data, calibration factors were derived from [ 177 Lu]Lu phantoms.

Ex Vivo Biodistribution
For the ex vivo biodistribution studies, animals were euthanized at selected time points (4,24,48, and 72 h) after injection. Specific tissues and tumors were excised, and their radioactivity uptake was determined. The following organs were collected from each animal: blood, tumor, heart, lung, liver, spleen, stomach, intestine, pancreas, kidney, muscle, skin, bone, and bone marrow. To confirm receptor specificity, mice were co-injected with [ 177 Lu]Lu-JR11 or [ 177 Lu]Lu-8a and a 50-molar excess of their respective unlabeled compound (JR11 or 8a), after which uptake in organs and tumor was determined at 24 h post-injection. All tissues were weighed and counted in a γ-counter, and data were reported as percentage injected dose per gram of tissue (% ID/g).

Statistical Analysis
Statistical analysis and nonlinear regression were performed using GraphPad Prism 9 (GraphPad software, San Diego, CA, USA), and a Mann-Whitney test was used to compare medians between groups. Data were reported as mean ± SEM (standard error of mean) for at least three independent replicates.

Conclusions
Our manuscript reported a successful synthesis of two long-acting JR11 analogs for improved radionuclide therapy of NETs. Radiolabeling of both analogs with lutetium-177 was achieved with very high RCYs and RCPs. Both radiopeptides showed excellent stability in PBS and mouse serum, conserved their hydrophilic behavior, and exhibited good binding to human albumin. Compounds 8a and 8b showed good binding affinity towards SSTR2, high cell uptake, and low internalization rate. [ 177 Lu]Lu-8a demonstrated extended residence in the blood, higher kidney uptake, and nonspecific tumor accumulation compared to [ 177 Lu]Lu-JR11. Unfortunately, [ 177 Lu]Lu-8b did not show any tumor uptake despite the high potential of the 4-(p-methoxyphenyl)butyryl ABD. Although insertion of a 4-(p-iodophenyl)butyryl ABD into JR11 improved its blood circulation, as expected, we also noticed high uptake in non-target organs with [ 177 Lu]Lu-8a. Therefore, further optimization is required to combine an ABD and JR11 to obtain a long-acting SSTR2 antagonist with an adequate biodistribution and pharmacokinetic profile for safe and efficient radionuclide therapy of neuroendocrine tumors.