Pyridyl-Ala in the third position of radiolabeled somatostatin antagonists: the effect of regioisomeric substitution

Abstract

Background

The radiolabeled somatostatin receptor subtype 2 (SST₂) antagonists LM3 (Phe(4-Cl)¹‐c(dCys²‐Tyr³‐dAph(Cbm)⁴‐Lys⁵‐Thr⁶‐Cys⁷)-dTyr⁸-NH₂) and JR11 (Phe(4-Cl)¹‐c(dCys²‐Aph(Hor)³‐dAph(Cbm)⁴‐Lys⁵‐Thr⁶‐Cys⁷)-dTyr⁸-NH₂) are under clinical evaluation for imaging and treatment of neuroendocrine tumors. These peptides differ at position 3, where LM3 contains Tyr³, while JR11 incorporates Aph(Hor)³. The amino acid at this position is crucial in the design of somatostatin ligands, agonists and antagonists, influencing affinity and receptor subtype specificity. Pyridylalanine, a nonnatural amino acid, presents three regioisomers 2-pyridylalanine (2Pal), 3-pyridylalanine (3Pal), and 4-pyridylalanine (4Pal), which differ only in the nitrogen atom’s position in the aromatic ring, allowing minimal chemical modification. We investigated whether the new somatostatin antagonists DOTA-[2Pal³]-LM3, DOTA-[3Pal³]-LM3 and DOTA-[4Pal³]-LM3, radiolabeled with Lu-177, differ among them and how they compare with the clinically used [¹⁷⁷Lu]Lu-DOTA-LM3.

Results

The synthesis of the DOTA-[2Pal³]-LM3 resulted in the formation of two diastereomers, with the d2Pal derivative lacking receptor recognition and affinity, contrary to the enantiomer l (l2Pal) derivative. The hydrophilicity of [¹⁷⁷Lu]Lu-DOTA-[xPal³]-LM3 increased in the order of l2Pal < 3Pal < 4Pal (logD = -2.3 ± 0.1 -2.5 ± 0.1 and -2.6 ± 0.1, respectively), being similar or significantly higher than [¹⁷⁷Lu]Lu-DOTA-LM3 (logD = -2.3 ± 0.1). Saturation binding studies indicated a trend of affinity improvement by l2Pal < 3Pal < 4Pal (K_D = 0.18 ± 0.02, 0.15 ± 0.01 and 0.11 ± 0.01 nM, respectively), which is similar to [¹⁷⁷Lu]Lu-DOTA-LM3 (K_D = 0.09 ± 0.02 nM). Surprisingly, despite similar accumulation in SST₂-positive tumors, differences were observed in the body distribution. The hydrophilicity of the Pal amino acids is likely responsible for the higher kidney uptake of the three ¹⁷⁷Lu-Pal-radioligands when compared to [¹⁷⁷Lu]Lu-DOTA-LM3. In particular, [¹⁷⁷Lu]Lu-DOTA-[3Pal³]-LM3 is characterized by high uptake and long retention in kidneys, probably due to its high stability in renal tissue. Chromatographic analysis of kidney homogenates revealed that more than 60% of peptide remained intact 1 h after injection.

Conclusions

Our study revealed that the replacement of Tyr³ with Pal³ isomers does not impact on SST₂ affinity, but chirality at this position is critical, as the d2Pal³ derivative loses binding. More interestingly, we demonstrated how the nitrogen’s position in the pyridylalanine regioisomers influences the properties of the corresponding radioligand. The polar nature of the 3Pal, due to its electronic density dissymmetry, likely enhances the peptide interaction with specific kidney transporters explaining its high uptake and prolonged retention in renal tissue.

Supplementary Information

The online version contains supplementary material available at 10.1186/s41181-025-00363-6.

Introduction

Radiolabeled somatostatin analogs play a pivotal role in the management of neuroendocrine tumor (NET) patients due to the high expression of somatostatin receptors (SST), especially the subtype 2 (SST₂), on NET cells (Ambrosini et al. 2021). Nowadays, the SST₂ agonists [⁶⁸Ga]Ga-DOTA-TOC (⁶⁸Ga-DOTA-[Tyr³]-octreotide, where DOTA = 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetrayl)tetraacetic acid) and [⁶⁸ Ga]Ga/[¹⁷⁷Lu]Lu-DOTA-TATE (⁶⁸Ga/¹⁷⁷Lu-DOTA-[Tyr³,Thr⁸]-octreotide) are approved and part of the clinical routine. These are analogs of octreotide (dPhe¹‐c(Cys²‐Phe³‐dTrp⁴-Lys⁵‐Thr⁶‐Cys⁷)-Tyr(ol)⁸), commonly substituted in position 3 by Tyr³. Over the last few years, radiolabeled octreotide-based SST₂ antagonists showed certain advantages over agonists and entered into clinical trials for imaging and treatment of NETs (Fani et al. 2017; Imperiale et al. 2023). Among these ligands, the SST₂ antagonists LM3 = Phe(4-Cl)¹‐c(dCys²‐Tyr³‐dAph(Cbm)⁴‐Lys⁵‐Thr⁶‐Cys⁷)-dTyr⁸-NH₂ (Fani et al. 2011), where dAph(Cbm) = D-4-amino-carbamoyl-phenylalanine, and JR11 = Phe(4-Cl)¹‐c(dCys²‐Aph(Hor)³‐dAph(Cbm)⁴‐Lys⁵‐Thr⁶‐Cys⁷)-dTyr⁸-NH₂ (Fani et al. 2012), where Aph(Hor) = 4-amino-L-hydroorotyl-phenylalanine, stand out due to their excellent targeting properties, i.e. high affinity and selectivity for SST₂, and optimal in vivo characteristics, such as high and persistent tumor uptake. As of today, DOTA-LM3 labeled with ¹⁷⁷Lu (Baum et al. 2021) or ¹⁶¹Tb (Fricke et al. 2024) and DOTA-JR11 labeled with ¹⁷⁷Lu (Wild et al. 2023), along with their corresponding NODAGA (1,4,7-triazacyclononane,1-glutaric acid-4,7-acetic acid) conjugates labeled with ⁶⁸Ga (Nicolas et al. 2018; Lin et al. 2023; Zhu et al. 2021) and ⁶¹Cu (NCT06455358), are in the focus of clinical development as new theranostics for NETs. Clinical data on radiolabeled somatostatin receptor antagonists indicate several advantages over currently used agonists, including higher and more prolonged tumor uptake, as well as increased imaging sensitivity (Imperiale et al. 2023; Wild 2024).

The third position of the octreotide structure is involved in the critical type II’ β-turn formed by the active core Phe³‐dTrp⁴-Lys⁵‐Thr⁶. Early on, modifications of this core, demonstrated the impact of position 3 on key properties, such as receptor affinity. For instance, substitution of Phe³ by Tyr³ significantly improved the affinity for SST₂ – a starting point in the development of DOTA-TOC and DOTA-TATE – while further modifications impacted in affinity and subtype selectivity (Ginj et al. 2006a, b; Fani et al. 2022). Among various natural and unnatural amino acids that were tested at this position, the unnatural amino acid pyridylalanine is particularly interesting for two main reasons: a) it presents three regioisomers, 2-pyridylalanine (2Pal), 3-pyridylalanine (3Pal), and 4-pyridylalanine (4Pal), which differ only in the position of the nitrogen atom in the aromatic ring, and b) 3Pal was reported to enhance antagonistic potency in peptide analogs of somatostatin and gonadotropin releasing hormone (Hocart et al. 1998, 1999; Rivier et al. 1986). Recently, LM4 (Phe(4-Cl)¹‐c(dCys²‐4Pal³‐dAph(Cbm)⁴‐Lys⁵‐Thr⁶‐Cys⁷)-dTyr⁸-NH₂), containing the unnatural amino acid 4-pyridilalanine (4Pal) in third position, has been conjugated to the chelators DATA and AAZTA, allowing faster coordination of the radiometals ⁶⁸Ga and ¹¹¹In/¹⁷⁷Lu under milder conditions, facilitating clinical application (Nock et al. 2023).

The aim of this study was to investigate the impact of the pyridylalanine regioisomers, as the minimal chemical modification, in position 3 of radiolabeled SST₂ antagonists. The LM3 structure was chosen for this purpose. Three new radioligands, [¹⁷⁷Lu]Lu-DOTA-[2Pal³]-LM3, [¹⁷⁷Lu]Lu-DOTA-[3Pal³]-LM3, and [¹⁷⁷Lu]Lu-DOTA-[4Pal³]-LM3 were synthesized and evaluated head-to-head with the reference ligand [¹⁷⁷Lu]Lu-DOTA-LM3. Their SST₂ binding characteristics were studied in vitro in HEK-SST₂ cells and membranes. Their in vivo evaluation was performed in HEK-SST₂ xenografts via SPECT/CT imaging, biodistribution and metabolic studies.

Materials and methods

The N^α-fuorenylmethoxycarbonyl (Fmoc)-protected amino acids and Rink Amide MBHA resin (0.56 mmol/g) were purchased from Merck Novabiochem (Darmstadt, Germany). DOTA(tert-butyl(tBu))₃ was obtained from CheMatech (Dijon, France). Fmoc-dAph(Cbm(tBu))-OH and Fmoc-2Pal-OH were obtained from Chem-Impex (Wood Dale, USA). Fmoc-Phe(4-Cl)-OH, ethyl cyanohydroxyiminoacetat (oxyma) and N,N′-diisopropylcarbodiimide (DIC) were obtained from Bachem (Bubendorf, Switzerland), CEM (Matthews, USA) and Iris biotech (Marktredwitz, Germany), respectively. [¹⁷⁷Lu]LuCl₃ (non-carrier added) was kindly provided by ITM (Munich, Germany). All other reagents and solvents were purchased from Acros Organics (Geel, Belgium), Merck Millipore (Darmstadt Germany) and Bioconcept (Allschwill, Switzerland), and used without further purification. Quantitative γ-counting was carried out on a COBRA 5003 γ-system well counter from Packard Instruments (Meriden, USA). Liquid chromatography mass spectrometry (LC–MS) was run on a LCMS-2020 SHIMADZU equipped with a Waters XBridge C-18 column (4.6 × 150 mm, 5 µm particle size); eluent A = H₂O (0.1% trifluoroacetic acid (TFA)), eluent B = Acetonitrile (0.1% TFA), gradient A = 10–50% eluent B in 15 min, flow rate: 1.0 mL/min, with the ultraviolet (UV) detector channel A set at a wavelength ($\lambda$) of 214 nm. For purification purposes, a semi-preparative Waters XBridge C-18 column (10 × 150 mm, 5 µm particle size) was used with a flow rate of 5 mL/min. For the quality control of the radioligands, high performance liquid chromatography (HPLC) was performed on the HPLC-Agilent system equipped with a Berthold LB509 γ-detector using a Proteo Jupiter C-12 column (4 × 250 mm, 4 µm particle size), gradient B = 0–50% eluent B in 15 min, flow rate: 1.0 mL/min.

Synthesis of the DOTA-conjugated peptides

DOTA-LM3, DOTA-[2Pal³]-LM3, DOTA-[3Pal³]-LM3 and DOTA-[4Pal³]-LM3 were assembled on the automated microwave peptide synthesizer Liberty Blue (CEM, Charlotte, NC, USA), following the Fmoc/tBu strategy and using Rink-Amide resin. Detailed synthetic procedure and purification are described in the Supplementary Information. All DOTA-conjugated ligands were obtained with a purity greater than 95%, as determined by HPLC.

Metalation and radiolabeling

The complexes with natural isotope of Lu (^natLu) were formed after incubation of the DOTA-conjugates with 2.5-fold excess of ^natLuCl₃ × 5H₂O in ammonium acetate buffer, 0.4 M, pH 5 at 95 °C for 30 min. Free metal ions (^natLu³⁺) were eliminated by SepPak C-18 purification cartridges (Waters), pre-conditioned with methanol and Milli-Q-water 1:1 v/v. Free ^natLu was eluted with Milli-Q-water, while the ^natLu-complexes were eluted with methanol. The methanol phase was evaporated, re-dissolved in water, and lyophilized. Quality control was performed with RP-HPLC and ESI–MS.

The ¹⁷⁷Lu-DOTA-conjugates were obtained by incubating 3 nmol of the corresponding DOTA-conjugate in 250 µL of ammonium acetate buffer (0.4 M, pH 5.0) with [¹⁷⁷Lu]LuCl₃ (40–120 MBq, depending on the planned experiment) for 30 min at 95 °C. For quality control, a small aliquot of the radiolabeled solution was withdrawn, mixed with 50 µL of a 0.1 M Ca-DTPA (calcium diethylenetriaminepentaacetic acid) solution, and subsequently analyzed by radio-RP-HPLC. Ca-DTPA forms a water-soluble complex with any unbound ¹⁷⁷Lu³⁺, preventing its interaction with or adsorption onto the HPLC column material and ensuring its complete elution. This enables accurate determination of the radiolabeling yield. For the saturation binding assay, an equivalent amount of ^natLuCl₃ × 5H₂O (3 × 10^–4 M) was added and the radiolabeling solution was further incubated for 30 min at 95 °C. This process is carried out to complex the non-labeled DOTA-conjugated molecules with ^natLu, thereby yielding a structurally homogeneous ligand solution composed of ¹⁷⁷Lu/^natLu-DOTA-conjugates. The reaction solution was then diluted with saline in order to obtain the final dilutions.

Lipophilicity

Lipophilicity was determined by the “shake-flask” method. Ten µL of the radioligand (0.1 µM) were added to a well-mixed mixture of 500 µL of octanol and 500 µL of phosphate-buffered saline (PBS), pH = 7.4, in triplicate. The solutions were vortexed for 3–5 min, mixed in a shaker for 1 h and centrifuged for 10 min at 3,000 rpm. From each phase three aliquots of 100 µL were withdrawn and measured in the γ-counter. The distribution coefficient log D_(pH=7.4) was calculated as the average log ratio value of the radioactivity in the organic fraction and PBS fraction, out of 2–3 independent experiments, each in triplicate. (log D_(pH=7.4) = Log₁₀ (counts octanol phase/counts aqueous phase)).

In vitro experiments

Cell culture

The human embryonic kidney HEK-293 cell line expressing the T7-epitope tagged human SST₂ receptor (HEK-SST₂) was provided by Prof. Stefan Schulz (Institute of Pharmacology and Toxicology, Jena University Hospital, Jena, Germany). HEK-SST₂ cells were cultured at 37 °C and 5% CO₂ in Dulbecco’s modified Eagle Medium (DMEM) containing 10% fetal bovine serum (FBS), 100 U/mL penicillin, 100 mg/mL streptomycin, 200 µmol/mL L-Glutamin, and 500 mg/mL G418. For all cell experiments, HEK-SST₂ cells were seeded at density of 1 × 10⁶ cells/well in 6-well plates and incubate overnight in DMEM with 1% FBS. For plating HEK-SST₂ cells, the plates were pre-treated with a solution of 10% (v/v) poly-L-lysine to promote the cell attachment.

All in vitro assays described below were performed with n = 2–3 independent experiments for each radioligand, with each experiment conducted in triplicate.

Internalization studies

On the day of the experiment, the cells were incubated with the radioligand (2.5 nM, 0.05–0.1 MBq) for 30, 60, 120 and 240 min at 37 °C. The non-specific cellular uptake was determined by adding 100 µL of NODAGA-JR11 (2.5 µM) in selected wells. The internalization was stopped by removing the medium and washing the cells twice with ice-cold PBS. This represented the free fraction. The cell surface-bound (acid releasable) fraction was collected with an acidic glycine solution (pH = 2.8) while the internalized fraction was obtained collecting the cells with a basic solution of NaOH (1 M). The specific cellular uptake (sum of membrane bound + internalized fractions), and the internalized and cell surface bound fractions separately, were quantified using a γ-counter and represented as a percentage of the total activity added per well.

Dissociation studies

For the dissociation studies the plated cells were first placed at 4 °C for 30 min and then incubated with the radioligand (2.5 nM, 0.05–0.1 MBq) and NODAGA-JR11 as blocking agent in selected well (2.5 µM). After 2 h, the free fraction was collected by removing the medium and washing twice with PBS, then the cells were incubated at 37 °C in presence of only fresh medium or with an additional 1,000-fold excess of ^natLu-DOTA-JR11 (used as competitor, 2.5 µM) for different time points (ranging between 10 and 240 min). At each time point, the medium was removed and replaced by fresh pre-warmed (37 °C) medium alone or containing the competitor. At the end of the experiment, the cells were detached with NaOH (1 M) and collected to determinate the remaining cell-associated radioligands. The different fractions were quantified using a γ-counter and represented as a percentage of the applied radioactivity.

Saturation binding assay

Saturation binding assays were conducted in HEK-SST₂ cell membranes, and the data were analyzed using GraphPad Prism 8 Software. The preparation of the cell membranes is described in the Supplementary Information. For the saturation binding assay, the membrane suspension (10 µg) was incubated with different concentrations (0.075–10 nM, 0.003–0.4 MBq) of each radioligand (30 µL) in binding buffer. To determinate non-specific binding, in selected wells, 1000-fold excess of NODAGA-JR11 (blocking agent) was added. After 1 h at 37 °C the mixture was harvested on a Brandel 48-well Cell Harvester using GF/C filter (Whatman), pre-soaked with 0.1% polyethyleneimine which helps the membranes to stick to the filter. The filters were washed with a HEPES buffer (50 mM) to remove the part of the radio conjugate not bound to the membrane. Finally, the filters were collected in different tubes and the activity measured using a γ-counter.

Inhibition assay

HEK-SST₂ cell membrane suspension, corresponding to 10 µg/well, was incubated for 1 h at 37 °C, with increased concentrations of each ^natLu complexed ligand (ranging between 0.1 and 1000 nM) in 96-wells plate using ¹²⁵I-Tyr-SS-14 (0.5 nM) as competitor. For the non-specific series, in three selected wells SS14 (1 µM) was added. Rapid filtration followed on a Brandel 48-well Cell Harvester using GF/C filter (Whatman), pre-soaked with 0.1% PEI solution. The filter was washed with 3 × 1 mL 20 mM HEPES. Filter activity was measured on an automatic gamma-counter. Non-specific binding is defined as the amount of activity binding in the presence of 1,000-fold excess of SS14. The data were analysed by GraphPad Prism 8 Software and the IC₅₀ values were determined using the ‘log(inhibitor) vs response’ equation (Y = Bottom + (Top–Bottom)/(1 + 10^(X-LogIC50))).

In vivo experiments

Animal model

Animal experiments were conducted in accordance with the Swiss animal welfare laws and regulations under license number 30515 granted by the Veterinary Office (Department of Health) of the Cantonal Basel-Stadt. Female athymic nude-Foxn1^nu/Foxn1⁺ mice (Envigo, The Netherland), 4–6 weeks old, were inoculated subcutaneously with 10⁷ HEK-SST₂ cells, freshly suspended in 100 µL sterile PBS, in the shoulder. The xenografted mice were monitored and the tumors were allowed to growth for 3–4 weeks, reaching a volume of 100–200 mm³.

SPECT/CT imaging

HEK-SST₂ xenografted mice were euthanized at 4 h after intravenous injection via the tail vein of the radioligand (200 pmol/100 µL/15 MBq) and imaged supine, head first, using a SPECT/CT system dedicated to small animals (NanoSPECT/CT™ Bioscan Inc.). A helical CT scan was acquired with the following parameters: current, 177 mA; voltage, 45 kVp; pitch, 1. A helical SPECT scan was acquired using multipurpose pinhole collimators (APT1), 20% energy window width centered symmetrically over the 208- and 113-keV γ-peaks of ¹⁷⁷Lu, 24 projections, and 1,200 s per projection. CT and SPECT images were reconstructed and filtered using the manufacturer’s algorithm, resulting in a pixel size of 0.3 mm for the SPECT and of 0.2 mm for the CT.

Biodistribution

Quantitative biodistribution studies were performed for all radioligands (10 pmol/100 μL/0.4–0.5 MBq) at 4 h and 24 h p.i. At the pre-selected time points, the mice (3–5 per group) were euthanized by CO₂ asphyxiation. Organs of interest and blood were collected, rinsed, blotted dry, weighed and counted in a γ-counter. The samples were counted against a suitably diluted aliquot of the injected solution as the standard. The results are expressed as percentage of injected activity per gram (%IA/g) and represent the mean ± standard deviation (SD).

In vivo metabolic stability

The in vivo stability was assessed in healthy BALB/c mice after intravenous injection via the tail vein of each radioligand (100 μL/500 pmol/ 8–9 MBq). The mice were euthanized by CO₂ and blood samples were collected at 30 min and 1 h p.i. in polypropylene tubes containing EDTA. The mixtures were vortexed, and subsequently centrifuged at 4,000 g for 10 min at 4 °C. Kidneys were quickly removed after euthanasia and were transferred to ice-cooled polypropylene tubes containing 1% ReadyShield® Protease Inhibitor Cocktail (Sigma-Aldrich) in PBS. The kidney tissues were homogenized using an Ultra-Turrax homogenizer, followed by centrifugation at 15,000 g for 15 min at 4 °C. The supernatant of each sample was collected, mixed with methanol (v/v 1:2) and centrifuged again for additional 15 min. Aliquots of the supernatant obtained from the blood and kidney homogenate were diluted 1:1 with H₂O up 1 mL and analyzed by radio-RP-HPLC on an Agilent 1260 Infinity system (see Reagents and Procedures above). The samples were analyzed on a Proteo Jupiter C12 (5 µm, 250 × 4.6 mm) column using the gradient B.

Results

Chemistry and radiochemistry

The chemical structures of the studied peptides conjugated to DOTA are shown in Fig. 1. The ligands were analyzed by reverse-phase high-performance liquid chromatography mass spectrometry (RP-HPLC–MS). Analytical data are summarized in Table S1. The RP-HPLC chromatograms are reported in Figure S1.

Fig. 1.

Chemical structures of the four DOTA-conjugated peptides evaluated in this study

During the synthesis of DOTA-[2Pal³]-LM3, LC–MS analysis of the crude material revealed the presence of two compounds, with very similar retention time and the same mass/charge (m/z) (768.4 and 768.1 u, respectively), indicating racemization (Figure S2). Both peaks were collected, purified and characterized. Chiral integrity of each peak was determined by C.A.T. GmbH (Tübingen, Germany), using deuteration followed by gas chromatography mass spectrometry (GC–MS). Briefly, the peptide was hydrolyzed in 6N DCl/D₂O whereby racemization was accompanied by proton-deuteron exchange in the α-C position (deuterium label). The enantiomers were gas chromatographically separated on chiral capillary using electron ionization single ion monitoring (EI SIM) mass spectrometry. The relative amounts of D and L enantiomers originally present in the sample (before hydrolysis) were determined by monitoring the non-deuterated molecular ions or suitable fragment ions of both enantiomers. The analysis verified that peak 1 corresponded to the conjugate with the l enantiomer of the 2Pal (l2Pal), i.e. DOTA-[l2Pal³]-LM3, containing 7.31% of d2Pal amino acid, and peak 2 corresponded to the conjugate with the d enantiomer of the 2Pal (d2Pal), i.e. DOTA-[d2Pal³]-LM3, containing 11.1% of the l2Pal. The ligands were further purified to eliminate their enantiomeric contaminant and used for radiolabeling after reaching purity > 95%. The HPLCs of the pure ligands are shown in the Supplementary Figure S2. The l/d indication for the 2Pal enantiomer is used for each conjugate, i.e. DOTA-[l2Pal³]-LM3 and DOTA-[d2Pal³]-LM3 for clarification purposes, while obviously the DOTA-[3Pal³]-LM3 and DOTA-[4Pal³]-LM3 refer to the l enantiomer.

The natural Lutetium (^natLu) complexes were obtained with a yield of 70–80%, based on the initial amount of peptide used. The purity and identity were confirmed by RP-HPLC–MS. ¹⁷⁷Lu-labeling did not require any purification step. Radiochemical yield (non-isolated product, estimated by radio-HPLC) was > 99% and radiochemical purity > 95% (Figure S3).

Log D determination

The Pal³ substitution for Tyr³ led to higher hydrophilicity, as indicated by the distribution coefficient between organic and aqueous phase of [¹⁷⁷Lu]Lu-DOTA-[3Pal³]-LM3 and [¹⁷⁷Lu]Lu-DOTA-[4Pal³]-LM3 (-2.51 ± 0.06 and -2.58 ± 0.08, respectively), in comparison to [¹⁷⁷Lu]Lu-DOTA-LM3 (-2.27 ± 0.06, p =  < 0.0001). Interestingly, the log D of the two ¹⁷⁷Lu-labeled diastereomers of the 2Pal derivative showed a significant difference, with the l-diastereomer being more lipophilic than the d-diastereomer (-2.23 ± 0.07 vs -2.58 ± 0.13, p < 0.0001).

In vitro evaluation

Cellular uptake

All the ¹⁷⁷Lu-labeled ligands, except for DOTA-[d2Pal³]-LM3, showed high and SST₂-mediated (SST₂-specific) cellular uptake (Fig. 2A). Their cell associated activity was characterized by accumulation on the cell surface (50% up to 70% of the added activity, at 4 h/37 °C, depending on the radioligand) and low level of internalization (around 15%, at 4 h/37 °C). Interestingly, [¹⁷⁷Lu]Lu-DOTA-[d2Pal³]-LM3 show neither accumulation on the cell membrane (< 2% at 4 h/37 °C), nor internalization (< 1% at 4 h/37 °C, Figure S2, panel C). Among the tested radioligands, [¹⁷⁷Lu]Lu-DOTA-[3Pal³]-LM3 and [¹⁷⁷Lu]Lu-DOTA-[4Pal³]-LM3 showed similar cellular uptake (82.90 ± 2.42% and 83.20 ± 0.70% of the added activity), being significantly higher compared to [¹⁷⁷Lu]Lu-DOTA-LM3 (67.25 ± 1.41%, p < 0.0001 for both). This was the same case when compared to [¹⁷⁷Lu]Lu-DOTA-[l2Pal³]-LM3 (77.12 ± 1.97%, p = 0.0012 and 0.0016, respectively). The time dependent cellular uptake and distribution for each radioligand is reported in Figure S4.

Fig. 2.

A Specific cellular uptake (cell surface bound + internalized fractions) for [¹⁷⁷Lu]Lu-DOTA-LM3 (orange), [¹⁷⁷Lu]Lu-DOTA-[l2Pal³]-LM3 (black), [¹⁷⁷Lu]Lu-DOTA-[3Pal³]-LM3 (blue) and [¹⁷⁷Lu]Lu-DOTA-[4Pal³]-LM3 (red), divided in specific cell surface bound fraction (dotted), and internalized fraction (full color). B Dissociation curves of tested radioligands with ^natLu-DOTA-JR11 as competitor. The data were analyzed according to one phase exponential decay equation (GraphPad Prism (version 8))

Dissociation

In the presence of a high affine SST₂ competitor, namely ^natLu-DOTA-JR11 (IC₅₀ = 0.73 ± 0.15 nM (Fani et al. 2012)), to avoid rebinding, all four radioligands dissociated following a one-phase exponential decay model (Fig. 2B). Among the tested radioligands, [¹⁷⁷Lu]Lu-DOTA-[l2Pal³]-LM3 dissociated faster than all others (K_off = 0.0179 ± 0.0004 min⁻¹ and T_1/2 = 38.73 min), while [¹⁷⁷Lu]Lu-DOTA-[4Pal³]-LM3 presented the strongest peptide-receptor interaction with the lowest k_off (0.0096 ± 0.0001 min⁻¹) and highest T_1/2 (71.85 min) (Table 1). The same trend was observed when dissociation was assessed in naïve medium without any competitor (Figure S5) with [¹⁷⁷Lu]Lu-DOTA-[4Pal³]-LM3 showing the highest residence time (T_1/2 = 87.5 min).

Table 1.

Total density of the receptor, affinities and dissociation parameters for all four ¹⁷⁷Lu-labeled ligands

	[177Lu]Lu-DOTA-LM3	[177Lu]Lu-DOTA-[l2Pal3]-LM3	[177Lu]Lu-DOTA-[3Pal3]-LM3	[177Lu]Lu-DOTA-[4Pal3]-LM3
Bmax (nM)	0.31 ± 0.01	0.47 ± 0.01	0.38 ± 0.01	0.39 ± 0.01
KD (nM)	0.09 ± 0.02	0.18 ± 0.02	0.15 ± 0.01	0.11 ± 0.01
IC50 (nM)	0.85 (0.7–1.0)	2.1(1.8–2.5)	2.0 (1.7–2.5)	0.80 (0.7–0.9)
t1/2 (min)	56.5 (52.1–61.8)	38.7 (36.5–41.2)	45.2 (41.0–50.5)	71.8 (69.3–74.6)
koff (min−1)	0.0123 ± 0.0004	0.0179 ± 0.0004	0.0153 ± 0.0006	0.0096 ± 0.0001

Total density of the receptors (B_max in nM), dissociation constant (K_D in nM), half maximal inhibitory concentration (IC₅₀ in nM), half-life (t_1/2) (min) and dissociation rate constant (k_off) (min⁻¹) of all the radioligands with ^natLu-DOTA-JR11 as competitor

Saturation binding assay

All the radioligands showed similar saturation binding profiles, with dissociation constant in a low nanomolar range. The K_D values ranged between 0.09 ± 0.02 nM ([¹⁷⁷Lu]Lu-DOTA-LM3) and 0.18 ± 0.02 nM ([¹⁷⁷Lu]Lu-DOTA-[l2Pal³]-LM3), as shown in Fig. 3A and reported in Table 1. The B_max values for the three Pal derivatives are similar to DOTA-LM3 suggesting that the pyridylalanine substitution and its regioisomers do not impact on the maximum number of binding sites recognized by these radioligands.

Fig. 3.

A Saturation binding curves of [¹⁷⁷Lu]Lu-DOTA-LM3 (orange), [¹⁷⁷Lu]Lu-DOTA-[l2Pal³]-LM3 (black), [¹⁷⁷Lu]LDOTA-[3Pal³]-LM3 (blue) and [¹⁷⁷Lu]Lu-DOTA-[4Pal³]-LM3 (red). The data were analyzed according to one site specific binding equation (GraphPad Prism (Version 8)). B Sigmoidal dose–response curves of the ^natLu-metaled conjugates in competition binding assays having ¹²⁵I-Tyr-SS14 as radioligand. The data were analyzed according to log(inhibitor) versus response equation (GraphPad Prism (Version 8))

Inhibition assay

IC₅₀ values, evaluated using ¹²⁵I-Tyr-SS-14 (0.05 nM) as a competitor, followed the same trend observed for the K_D, with ^natLu-DOTA-[4Pal³]-LM3 having similar IC₅₀ to ^natLu-DOTA-LM3, while ^natLu-DOTA-[l2Pal³]-LM3 and ^natLu-DOTA-[3Pal³]-LM3 showed 2.5-fold higher IC₅₀ values (Fig. 3B, Table 1). Nevertheless, all the ^natLu-DOTA-conjugates showed a low nanomolar affinity in agreement with the published affinity values of high affine somatostatin antagonist, such as ^natLu-DOTA-JR11 (IC₅₀ = 0.73 ± 0.15 nM (Fani et al. 2012)).

In vivo studies

SPECT/CT imaging

SPECT/CT images were acquired 4 h after intravenous injection of each radioligand in HEK-SST₂ xenografts and are shown in Fig. 4A. All radioligands were able to visualize and delineate the tumors demonstrating high in vivo affinity for the SST₂-positive tumors. The specificity of the Pal derivatives was evaluated for the [¹⁷⁷Lu]Lu-DOTA-[l2Pal³]-LM3 derivative, by injecting 1,000-fold excess of DOTA-LM3 3–5 min before the injection of the radioligand. SPECT/CT image was acquired 4 h p.i. (Fig. 4B).

Fig. 4.

A SPECT/CT images in SST₂-expressing tumor mouse model, 4 h after injection of 200 pmol/14–15 MBq of the corresponding radioligand. B SPECT/CT image in SST₂-expressing tumor mouse model, 4 h after injection of 200 pmol/14 MBq of [¹⁷⁷Lu]Lu-DOTA-[l2Pal³]-LM3 and pre-injection of 1,000-fold excess of DOTA-LM3

Biodistribution studies

Biodistribution studies were performed in HEK-SST₂ xenografts at 4 h and 24 h p.i.. The data revealed an overall similar biodistribution profile, besides the accumulation in the kidneys, where regioisomerization impacted significantly (Table 2). The first observation was the significantly higher kidney uptake of [¹⁷⁷Lu]Lu-DOTA-[3Pal³]-LM3, among all tested radioligands, at both time points of investigation. The kidney uptake was statistically significantly higher for [¹⁷⁷Lu]Lu-DOTA-[3Pal³]-LM3 in comparison to [¹⁷⁷Lu]Lu-DOTA-LM3 (18.8%IA/g vs 7.9%IA/g, p = 0.0011), [¹⁷⁷Lu]Lu-DOTA-[l2Pal³]-LM3 ( vs 11.0%IA/g, p = 0.0055) and [¹⁷⁷Lu]Lu-DOTA-[4Pal³]-LM3 (vs 8.6%IA/g, p = 0.0012) at 4 h p.i., and to [¹⁷⁷Lu]Lu-DOTA-LM3 (15.3 ± 2.87 vs 4.4%IA/g, p = 0.0003), [¹⁷⁷Lu]Lu-DOTA-[l2Pal³]-LM3 (vs 8.6%IA/g, p = 0.0041) and [¹⁷⁷Lu]Lu-DOTA-[4Pal³]-LM3 (vs 8.4%IA/g p = 0.003) at 24 h p.i.. The second observation was the impact of the 3Pal and 4Pal on the retention of the radioligands in the kidneys. More specifically, while washout from the kidneys between 4 and 24 h was observed for [¹⁷⁷Lu]Lu-DOTA-LM3 (from 7.91 ± 1.24 to 4.41 ± 0.42%IA/g, respectively p = 0.017) and [¹⁷⁷Lu]Lu-DOTA-[l2Pal³]-LM3 (from 11.0 ± 1.17 to 8.59 ± 0.84%IA/g, respectively, p = 0.016), the uptake remained constant for [¹⁷⁷Lu]Lu-DOTA-[3Pal³]-LM3 and [¹⁷⁷Lu]Lu-DOTA-[4Pal³]-LM3 between 4 and 24 h (18.8 ± 3.47 and 15.3 ± 2.87%IA/g, respectively, p = 0.18 and 8.63 ± 0.62 and 8.36 ± 0.48%IA/g, respectively, p = 0.53). All the radioligands showed a similar and distinct accumulation in the tumor at 4 h p.i. which remained constant at 24 h p.i.. Radioactive accumulation in receptor organs such as the pancreas, stomach, and intestine was high at the early time point of the study but washed out to a high degree, resulting in an improved tumor-to-organ ratios over time. The high activity observed in the lungs at the early time point of the study was mainly attributed to the blood content, as the lungs are highly vascularized.

Table 2.

Biodistribution data of 10 pmol and 0.4 MBq of the corresponding radioligands

Organ	[177Lu]Lu-DOTA-LM3		[177Lu]Lu-DOTA-[l2Pal3]-LM3		[177Lu]Lu-DOTA-[3Pal3]-LM3		[177Lu]Lu-DOTA-[4Pal3]-LM3
	4 h	24 h	4 h	24 h	4 h	24 h	4 h	24 h
Blood	0.08 ± 0.02	0.03 ± 0.00	0.08 ± 0.00	0.02 ± 0.00	0.10 ± 0.02	0.02 ± 0.00	0.08 ± 0.00	0.02 ± 0.00
Heart	0.50 ± 0.16	0.14 ± 0.02	0.35 ± 0.10	0.10 ± 0.02	0.34 ± 0.08	0.1 ± 0.03	0.41 ± 0.04	0.11 ± 0.02
Lung	10.3 ± 1.16	4.81 ± 1.74	7.10 ± 1.02	1.50 ± 0.13	7.43 ± 1.05	3.54 ± 1.58	7.31 ± 1.19	2.17 ± 0.41
Liver	1.78 ± 0.44	0.64 ± 0.11	1.84 ± 0.53	0.51 ± 0.11	1.60 ± 0.55	0.49 ± 0.1	1.23 ± 0.33	0.72 ± 0.16
Pancreas*	50.3 ± 4.00	24.9 ± 3.60	52.6 ± 5.26	14.6 ± 0.88	45.2 ± 7.17	15.0 ± 3.81	53.3 ± 3.55	21.6 ± 2.84
Spleen	0.97 ± 0.14	0.50 ± 0.13	0.80 ± 0.14	0.33 ± 0.03	0.86 ± 0.08	0.43 ± 0.07	0.75 ± 0.18	0.34 ± 0.03
Stomach*	42.4 ± 3.66	19.1 ± 1.15	36.0 ± 5.91	11.9 ± 1.68	34.6 ± 0.77	12.8 ± 0.72	37.9 ± 4.35	19.6 ± 3.48
Intestine*	12.0 ± 3.68	4.29 ± 1.35	9.54 ± 1.76	2.72 ± 0.66	9.92 ± 3.57	4.05 ± 0.99	9.64 ± 1.98	4.56 ± 0.89
Adrenal*	6.31 ± 2.07	3.61 ± 1.25	6.25 ± 1.19	2.73 ± 0.52	6.58 ± 1.89	2.29 ± 0.74	6.82 ± 0.51	3.21 ± 0.83
Kidney	7.91 ± 1.24	4.41 ± 0.42	11.0 ± 1.17	8.59 ± 0.84	18.8 ± 3.47	15.3 ± 2.87	8.63 ± 0.62	8.36 ± 0.48
Muscle	0.11 ± 0.04	0.03 ± 0.00	0.09 ± 0.00	0.03 ± 0.00	0.11 ± 0.03	0.03 ± 0.00	0.12 ± 0.05	0.03 ± 0.01
Femur	1.17 ± 0.31	0.85 ± 0.38	0.62 ± 0.23	0.24 ± 0.06	0.87 ± 0.14	0.60 ± 0.28	0.63 ± 0.13	0.44 ± 0.21
SST2-tumor	13.0 ± 2.01	12.3 ± 1.57	14.2 ± 0.90	15.0 ± 1.95	15.3 ± 5.17	14.5 ± 2.22	10.9 ± 0.90	13.2 ± 0.95
Tumor/	Ratios
Blood	163	410	177	750	153	725	136	660
Liver	7.3	19.2	7.7	29.4	9.6	29.6	8.9	18.3
Kidney	1.6	2.8	1.3	1.7	0.8	0.9	1.3	1.6
Muscle	118	410	158	500	139	483	91	440

The results are expressed as %IA/g ± SD (n = 3–5/group). * Mainly reported receptor-positive organs

In vivo metabolic stability

All four radioligands showed high stability in the blood with > 95% intact peptide 1 h after injection. However, the radioligands showed distinct differences in terms of their radiochemical forms in the kidneys, as revealed by chromatographic analysis at 30 min and 1 h after injection. [¹⁷⁷Lu]Lu-DOTA-[3Pal³]-LM3 showed the highest percentage of peptide remaining intact 1 h after injection (60%) (Fig. 5). At the same time, less than 50% of intact peptide was found for the other three radioligands, with [¹⁷⁷Lu]Lu-DOTA-[l2Pal³]-LM3 showing the fastest degradation (only 32% of the peptide was found intact in the kidneys at 1 h p.i.).

Fig. 5.

Radio-RP-HPLC chromatograms of the in vivo metabolic stability study in kidney homogenates of the four radioligands at 30 and 60 min p.i. The reported % refers to the intact radioligand

Discussion

The use of radiolabeled somatostatin receptor antagonists is continuously evolving, offering new options for the treatment of neuroendocrine tumors. Thus, the structural features that may impact on their in vivo distribution and/or pharmacokinetics are of particular interest. Like radiolabeled somatostatin receptor agonists, the position 3 in octapeptide sequence of the antagonists (all peptide sequences are based on octreotide, bearing Phe in this position) showed to be critical. However, most of these studies were conducted in vitro (Cescato et al. 2008; Hocart et al. 1999, 1998). Furthermore, limited data are available regarding the impact of the modifications at position 3 on the in vitro and in vivo properties of the radiolabeled, chelated versions of these antagonists. For example, [¹⁷⁷Lu]Lu-DOTA-JR11 and [¹⁷⁷Lu]Lu-DOTA-LM3, despite incorporating different amino acids at position 3 (Aph(Hor) and Tyr, respectively), exhibit very similar targeting propertie (e.g. IC₅₀, K_D, K_on/off and B_max), as reported by Mansi et al. (Mansi et al. 2021) for [¹⁷⁷Lu]Lu-DOTA-JR11, and in Table 1 of the present study for [¹⁷⁷Lu]Lu-DOTA-LM3. In this study, we investigated the impact of the minimal modification at this position by introducing the three regioisomers, 2Pal, 3Pal and 4Pal of the non-natural amino acid pyridylalanine in the sequence of the SST₂-antagonist DOTA-LM3.

The different position of the nitrogen in the pyridylalanine regioisomers influences its physico-chemical properties. The pyridine ring has a lower electron density than the phenyl ring and possesses hydrogen bonding capacity due to the negative character of the nitrogen in the aromatic ring. This, combined with the formation of a strong dipole, enhances the hydrophilicity of the peptides bearing the pyridine ring, compared to the natural amino acid Phe (Mroz et al. 2016). The 3Pal and 4Pal differ in charge distribution; 3Pal's meta position creates a dissymmetry in electronic density which is not the case in 4Pal's para position. The nitrogen in the ortho position of the 2Pal forms an intramolecular hydrogen bond with the peptide chain’s NH group, an interaction known as ‘masking of polarity,’ reducing in this case hydrophilicity (Vorherr et al. 2020). The racemization which occurred during coupling of the Fmoc-2Pal-OH was accompanied by a characteristic dark blue-brown coloration of the coupling solution after the addition of the two activators (DIC, oxyma) and it is reported in several peptide sequences, such as angiotensin II (Hsieh et al. 1979), luteinizing hormone releasing hormone (LHRH) (Folkers et al. 1983) and somatostatin analogs (Hocart et al. 1999). The two diastereomers, DOTA-l2Pal-LM3 and DOTA-d2Pal-LM3 showed different properties when radiolabeled with Lu-177. The internalization assay indicates that the change of amino acid (Pal vs Tyr) in position 3 does not affect the receptor recognition, while this is not the case regarding chirality, as the d diastereomer of the 2Pal showed no cellular uptake. This finding is in agreement with previous published data on somatostatin analogs reporting on the loss of affinity due to conformational changes from l to d of the same amino acid in position 3 (Cescato et al. 2008; Hocart et al. 1999). More specifically, Hocart et al (1999), hypothesized that the reversed spatial orientation of the 2Pal in d-conformation leads to a complete inability of the peptide to interact with the receptor. This hypothesis is further supported by Vorherr et al. (2020), who investigated the role of pyridylalanine modification in a cyclic hexapeptide. Their molecular design and NMR studies demonstrated that the diastereomeric change of the amino acid 2Pal between d and l forms leads to an inverted structure.

Despite the fact that [¹⁷⁷Lu]Lu-DOTA-LM3 is under clinical evaluation, there are only limited preclinical data published. [¹⁷⁷Lu]Lu-DOTA-LM3 is reported to have high and specific cellular uptake in the SST₂-positive AR42J cells (approx. 70% after 4 h) (Borgna et al. 2021), similar to our results in HEK-SST₂ cells (67.25 ± 1.41% after 4 h). Recently, [¹⁷⁷Lu]Lu-DOTA-LM3 has been evaluated in HEK-SST₂ transfected cells in comparison with the new radioligand [¹⁷⁷Lu]Lu-AAZTA⁵-LM4 (where the LM4 corresponds to the peptide sequence reported herein as [4Pal³]-LM3) (Nock et al. 2023). It is worth mentioning that in this work [¹⁷⁷Lu]Lu-DOTA-LM3 showed a cellular uptake very similar to our data (61.9 ± 3.0% vs 59.0 ± 3.0% of added activity at 1 h, Figure S4), while the [¹⁷⁷Lu]Lu-AAZTA⁵-LM4 presented 1.5-times lower cellular uptake compared to [¹⁷⁷Lu]Lu-DOTA-[4Pal³]-LM3 (47.7 ± 1.6% vs 72.6 ± 2.8% of added activity, respectively). This confirms the reported sensitivity of this class of radioligands to N-terminal modification (Fani et al. 2012).

Among the Pal derivatives, [¹⁷⁷Lu]Lu-DOTA-[l2Pal³]-LM3 and [¹⁷⁷Lu]Lu-DOTA-[3Pal³]-LM3 exhibited a higher dissociation constant (lower affinity), which is attributed to their faster dissociation (k_off), indicating a weaker receptor-ligand interaction, in comparison to [¹⁷⁷Lu]Lu-DOTA-[4Pal³]-LM3. Both, [¹⁷⁷Lu]Lu-DOTA-[l2Pal³]-LM3 and [¹⁷⁷Lu]Lu-DOTA-[3Pal³]-LM3 had similar dissociation properties (K_D and K_off) to [¹⁷⁷Lu]Lu-DOTA-JR11 studied under the same experimental conditions (Mansi et al. 2021). Contrary, [¹⁷⁷Lu]Lu-DOTA-[4Pal³]-LM3 had the lowest dissociation (highest cell residence time) among the Pal-derivatives, followed by [¹⁷⁷Lu]Lu-DOTA-LM3 (Table 1). All the radioligands showed high affinity in low nanomolar range, with the IC₅₀ of [¹⁷⁷Lu]Lu-DOTA-[4Pal³]-LM3 closely matching that of [¹⁷⁷Lu]Lu-DOTA-LM3. The affinity of [¹⁷⁷Lu]Lu-DOTA-LM3 for SST₂ previously determined using intact AR42J cells was lower (K_D = 1.6 ± 0.3 nM) compared to our data using HEK-SST₂ membranes. The different cell lines (transfected versus natural expressing) and the different experimental conditions (membrane versus intact cells) can explain the observed discrepancies (Borgna et al. 2021).

The main difference in vivo between [¹⁷⁷Lu]Lu-DOTA-LM3 and the ¹⁷⁷Lu-Pal-radioligands is their higher kidney uptake, clearly visible in the SPECT/CT images at 4 h p.i.; particularly pronounced for the 3Pal-derivative. As previously mentioned, the pyridine ring in the Pal derivatives is more hydrophilic than the phenyl ring in the LM3, which accounts for their increased accumulation in the kidneys and possibly their retention. Among the Pal derivatives, [¹⁷⁷Lu]Lu-DOTA-[3Pal³]-LM3 showed the highest kidney uptake, probably due to the asymmetry in the electronic density of its aromatic ring, which imparts a polar nature to the molecule. This distinct charge distribution may influence its pharmacokinetic behaviour, including its tendency to accumulate in the kidneys (de Roode, Joosten, and Behe 2024). Polar molecules are known to interact with renal transporters, such as organic cation transporters (OCTs) and/or organic anion transporters (OATs) facilitating their renal uptake and retention (Ullrich et al. 1993). This is in agreement with the observations in the in vivo metabolic stability study. More than 60% of [¹⁷⁷Lu]Lu-DOTA-[3Pal³]-LM3 was present in the kidney tissue in its intact form at 1 h p.i., indicating increased excretion. Furthermore, the higher percentage of intact radioligand found for the [¹⁷⁷Lu]Lu-DOTA-[3Pal³]-LM3, in comparison to the other radioligands, might also be due to increased stability in the kidneys, especially when considering that all radioligands showed the same high stability in the blood.

In contrast to the biodistribution data, accumulation in the abdominal organs is not visible in the images, most probably due to the injection of a higher peptide mass for imaging purposes (10 pmol vs 200 pmol). The results suggest a saturation effect that results in a faster saturation of receptors expressed in the positive organs compared to the tumor. This is in agreement with the findings regarding the “mass effect” in the case of the SST₂ antagonist [¹⁷⁷Lu]Lu-DOTA-JR11 (10 pmol vs 200 pmol) in the same tumor model (Nicolas et al. 2017) and the [¹⁷⁷Lu]Lu-DOTA-LM3 (40 pmol vs 200 pmol) in the AR42J xenografted model (Borgna et al. 2022). The mass effect was also evident in the uptake in the lungs, most probably due to their high blood content, reflecting the lower amount of radioligand in the blood pool when higher doses were administered (Borgna et al. 2022; Nicolas et al. 2017). The radioligands are characterized by a persistent accumulation in SST₂ tumors over time and by high uptake in the receptor positive organs, such as pancreas, stomach, intestine and adrenals at 4 h p.i. The higher accumulation of the radiolabeled somatostatin antagonists in the receptor positive organs, in comparison to the agonists, is a distinct difference in the biodistribution in mice between agonists and antagonists and has been reported previously by us and others (Borgna et al. 2022; Nicolas et al. 2017). This difference is mainly due to the higher number of binding sites recognized by the antagonists (Ginj et al. 2006b). However, the faster washout from these organs, compared to the tumor, led to an improved tumor-to-organ ratios over time. This improvement is more pronounced for [¹⁷⁷Lu]Lu-DOTA-[l2Pal³]-LM3 where tumor-to-pancreas and tumor-to-stomach ratios are 1.03 and 1.26 at 24 h, respectively. Nock et al.(Nock et al. 2023) reported that [¹⁷⁷Lu]Lu-AAZTA⁵-LM4 exhibited lower accumulation in SST₂-positive organs compared to [¹⁷⁷Lu]Lu-DOTA-[4Pal³]-LM3 when studied in HEK-SST₂ xenografts. This finding indicates a pronounced influence of the chelate system (Lu-DOTA vs Lu-AAZTA) on the biodistribution profile of the two radioligands and confirms the high sensitivity of this class of radioligands to chemical and structural modification. Interestingly, the same study reported distinct differences in the uptake of [¹⁷⁷Lu]Lu-DOTA-LM3 in SST₂-positive organs compared to our data (Nock et al. 2023). While the accumulation in the pancreas is very similar, our biodistribution data show approximately 3- and 4-times higher uptakes in the stomach and intestine, respectively. These differences may be attributed to the injection of a different peptide mass (10 pmol vs 40 pmol) and/or the use of a different mouse strain (Fox^nu vs SCID).

Conclusion

In this study, we elucidated the importance of the third position in octreotide-based somatostatin receptor antagonists by assessing the impact of the least possible chemical modification, i.e. regioisomerization via the three pyridylalanines, 2Pal, 3Pal, and 4Pal. Our study revealed that the replacement of Tyr³ in the [¹⁷⁷Lu]Lu-DOTA-LM3 with the Pal³ isomers does not impact on SST₂ affinity. However, the chirality at this position is critical, as shown by the loss of binding for the d2Pal³ derivative, in comparison to the l2Pal³, 3Pal³ and 4Pal³. The similar affinity and receptor recognition resulted in comparable uptake in SST₂-positive organs and tumors, but this study highlights the significant impact of the 3Pal amino acid on the high and persistent kidney uptake. This effect is driven by its polar nature and interaction with specific kidney transporters, ultimately influencing the overall pharmacokinetics of this radioligand.

Statistical analysis

The statistical analysis was performed using the program GraphPad Prism (Version 8). The p-values were calculated via GraphPad using a t test unpaired analysis, they were considered statistically significant when p < 0.05.

Author contributions

R.M., M.F. and H.M. conceived the project. R.M. designed the experiments and supervised N.B.. R.M., N.B., L.D.P., S.Z. and T.B.B. performed the experiments. R.M. and N.B. analysed the data. R.M. wrote the manuscript. M.F. edited the manuscript. All authors read and approved the final manuscript.

Funding

Open access funding provided by University of Basel. No specific funding was requested for this work.

Availability of data and material

All data generated or analysed during this study are included in this published article and its supplementary information file.

Declarations

Ethics approval and consent to participate

Animal experiments were conducted in accordance with the Swiss animal welfare laws and regulations under license number 30515 granted by the Veterinary Office (Department of Health) of the Cantonal Basel-Stadt.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.