The heterobivalent (SSTR2/albumin) radioligand [67Cu]Cu-NODAGA-cLAB4-TATE enables efficient somatostatin receptor radionuclide theranostics

Somatostatin type 2 receptor (SSTR2) radionuclide therapy using β- particle-emitting radioligands has entered clinical practice for the treatment of neuroendocrine neoplasms (NENs). Despite the initial success of [177Lu]Lu‑DOTA-TATE, theranostic SSTR2 radioligands require improved pharmacokinetics and enhanced compatibility with alternative radionuclides. Consequently, this study evaluates the pharmacokinetic effects of the albumin-binding domain cLAB4 on theranostic performance of copper‑67-labeled NODAGA-TATE variants in an SSTR2-positive mouse pheochromocytoma (MPC) model. Methods: Binding, uptake, and release of radioligands as well as growth-inhibiting effects were characterized in cells grown as monolayers and spheroids. Tissue pharmacokinetics, absorbed tumor doses, and projected human organ doses were determined from quantitative SPECT imaging in a subcutaneous tumor allograft mouse model. Treatment effects on tumor growth, leukocyte numbers, and renal albumin excretion were assessed. Results: Both copper‑64- and copper‑67-labeled versions of NODAGA-TATE and NODAGA-cLAB4‑TATE showed similar SSTR2 binding affinity, but faster release from tumor cells compared to the clinical reference [177Lu]Lu‑DOTA-TATE. The bifunctional SSTR2/albumin-binding radioligand [67Cu]Cu‑NODAGA-cLAB4‑TATE showed both an improved uptake and prolonged residence time in tumors resulting in equivalent treatment efficacy to [177Lu]Lu‑DOTA-TATE. Absorbed doses were well tolerated in terms of leukocyte counts and kidney function. Conclusion: This preclinical study demonstrates therapeutic efficacy of [67Cu]Cu‑NODAGA-cLAB4‑TATE in SSTR2-positive tumors. As an intrinsic radionuclide theranostic agent, the radioligand provides stable radiocopper complexes and high sensitivity in SPECT imaging for prospective determination and monitoring of therapeutic doses in vivo. Beyond that, copper‑64- and copper‑61-labeled versions offer possibilities for pre- and post-therapeutic PET. Therefore, NODAGA-cLAB4-TATE has the potential to advance clinical use of radiocopper in SSTR2-targeted cancer theranostics.


I. Radiolabeling of NODAGA-TATE variants with copper-67
Radiolabeling efficiency and peptide contents within the radioligand preparations Radiolabeling of NODAGA-TATE variants in sodium ascorbate buffer enables straightforward and efficient complexation of copper-67 even under mild conditions (40°C).Without additional purification, the achieved radiochemical purities of > 96 % are comparable to what is achieved in routine radiolabeling procedures for [ 177 Lu]Lu-DOTA-TATE at 80°C.This confirms that, due to its high molar activity and chemical purity, photonuclear reaction-produced [ 67 Cu]CuCl2 is well suited for radiolabeling of NODAGA-functionalized peptides.Notably, the remaining non-labeled precursor fraction was explicitly not saturated with the corresponding non-radioactive metal ions (in natural isotope composition) to match typical formulations of most clinical SSTR2 radioligand preparations [37,70].Therefore, the ratios between the distinct versions of the SSTR2 ligand within the radioligand preparation (radiometal complex, non-radioactive metal complexes, metal-free precursor) can be expected to modulate the pharmacokinetics to a certain extent.

Method: Cohort assembly and allocation to treatments
Mice with tumors of 6 ± 2 mm in diameter were grouped according to a similar average tumor volume per treatment group.The allocation to therapy and control groups was random.Consecutive experiments were performed according to the identical treatment protocol (Table S 1).

TATE variants in intact cells and cell homogenates
The development of novel SSTR2 ligands entails the characterization of their binding constants, which is often performed with radioligands in a saturation or competition binding assay format.Since SSTR2 is a G-protein-coupled transmembrane protein, receptor samples for binding assays are typically prepared from tissue sections, cell membranes, cell homogenates, or intact cells [34,[71][72][73].In the present study, binding constants measured in a microplate assay with intact MPC cells were compared with those measured in an MPC cell homogenate assay to examine whether functional cellular responses (e.g. receptor-mediated endocytosis, endosomal/lysosomal receptor traffic) may influence the results.

Binding constants measured with intact MPC cells and cell homogenates
The trends in Kd values and Bmax values within the series of gallium-68-, lutetium-177-, copper-64-, and copper-67-labeled-labled TATE variants were largely consistent between the two binding assays using intact MPC cells or cell homogenates (Figure S 1).However, Bmax values measured with the intact cell assay tend to be higher compared to the homogenate assay.This could suggest that, in intact cells, receptor-mediated endocytosis followed by endosomal recycling provides recurrent availability of SSTR2 on the cell surface without the concomitant stoichiometric release of radioligand, enabling repeated ligand binding by the same receptor molecule.Therefore, Bmax values measured for different radioligands in intact MPC cells not only depend on the number of SSTR binding sites but also on their uptake and recycling rates.Further general aspects to be considered in binding assays with intact cells or cell homogenates have been discussed elsewhere [34].

Hypotheses on lower binding capacities of MPC cells for copper-64-and copper-67-labeled NODAGA-TATE variants compared to gallium-68-and lutetium-177-labeled versions of DOTA-TATE
In binding studies performed with intact cells, radioligand-specific uptake ratios as well as rate constants for intracellular SSTR2 traffic (lysosomal trapping versus endosomal recycling) have been reported to functionally affect the measurement of Bmax values in vitro [45,46].For example, intact cells exhibit higher binding capacities for the SSTR2

IV. Inhibitory constants of metal-free, lutetium-, and copper-labeled TATE variants
Small structural modifications, chelator substitution or metal replacement have been reported to considerably affect the binding affinity of SSTR2 targeting peptide-chelator conjugates including agonists and antagonist [71,73,77].As shown in these studies, the effects on binding affinity resulting from changes in the metal-chelate complex can vary considerably between conjugate types.In the present study, it can be assumed that different ligand versions in the preparations of [ 67 Cu]Cu-NODAGA-TATE, [ 67 Cu]Cu-NODAGA-cLAB4-TATE and [ 177 Lu]Lu-DOTA-TATE (radiometal complex, non-radioactive metal complexes, metal-free precursor) modulate their pharmacological behavior to a certain extent.
Therefore, comparing the inhibitory constants of the corresponding non-radioactive reference ligands (metal-free and metalated versions) may help to estimate their mutual interference in vitro and in vivo.

Method: Radioligand uptake in tumor spheroids
Spheroids were treated with radioligands as described above.Total uptake of radioligands was determined for exposure at the highest initial activity concentration of 400 kBq/mL.Non-specific binding was determined in presence of 1 µmol acetyl-TATE.Binding to plastic surfaces was determined in spheroid-free cavities.Spheroids were soaked into UniFilter-96 GF/C microplates (PerkinElmer) and washed three times with Dulbecco's phosphate-buffered saline containing 0.9 mmol/L CaCl2 and 0.5 mmol/L MgCl2 using the UniFilter-96 Cell harvester (PerkinElmer).Activity of filters was measured using the gamma counter Wizard 2 2480 (PerkinElmer).Activity standards were prepared from serially diluted incubation media.Specific binding was calculated as percent initial dose normalized to spheroid volume (% ID/mm 3 ).(Am) molar activity at treatment start, (IAC) initial activity concentrations between 20-400 kBq/mL corresponding to initial doses (ID) between 4−80 kBq/per microwell followed by radioligand removal; (OC) overall control defined as < 2.2-fold change in spheroid circular area and (PFC) progression-free control defined as < 1.2-fold change in spheroid circular area compared to treatment start, respectively; (RN) radionuclide; (SCD50) half-maximal spheroid control doses are defined as the IAC providing a progression-free control for 7 days or b overall control for 14 days, in 50 % of spheroids; 32 spheroids were monitored for each treatment condition, uptake and SCD50 are presented as means ± standard error, PFC and OC are presented as medians

VI. SPECT imaging: acquisition, reconstruction, resolution, and activity calibration
Absolute quantification in preclinical single-photon emission computed tomography (SPECT) imaging provides a valuable non-invasive alternative to terminal biodistribution studies for monitoring the activity profiles of photon-emitting radioactive probes in small laboratory animals providing the basis for drug development, organ dosimetry, and therapy response assessment [81].Hence, the quantitative accuracy in SPECT imaging is essential for reliably assessing tumor uptake and tumor-to-normal tissue ratios necessary to perform dosimetry [82].Accurate image quantitation in small animal SPECT is always challenging due to the limitations in the instrumentation and imaging process; therefore, it is important to evaluate performance parameters of preclinical SPECT systems used for theranostic radionuclides during dosimetry studies [82].In the present study, SPECT images of lutetium-177-and copper-67-loaded activity phantoms were analyzed to evaluate radionuclide-specific count rates, spatial resolution, and quantifications accuracy.

Method: Image acquisition and reconstruction
Small animal single-photon emission computed tomography was performed using the nanoScan® SPECT/CT (Mediso Medical Imaging Systems, Budapest, Hungary) equipped with the APT56 aperture consisting of four M 3 multi-pinhole collimators providing a 30×30 mm axial field of view (FOV).Photon emission was recorded using frame times between 60 and 240 s (scan times between 20 and 80 min) and binned within the 20 % energy windows of the radionuclide-specific photopeaks for lutetium-177 ( 56

Figure S 1 :
Figure S 1: Binding of SSTR2 gallium-68-, lutetium-177-, copper-64-, and copper-67-labeled labeled TATE variants in different MPC cell preparations; (A) Total and non-specific binding towards intact cell monolayer; (B) Total and non-specific binding towards cell homogenates; (continuous line) total binding; (dashed line) non-specific binding; (C) Comparison of binding constants measured with intact cells and cell homogenates, respectively; binding was measured at free radioligand concentrations between 0.3−40 nmol/L; non-specific binding was assessed in presence of 1 nmol/L acetyl-TATE; (Am) molar activity at incubation start; number of independent experiments (n) each performed in a triplicate or b duplicate; data presented as means ± standard error; significance of difference compared to [ 177 Lu]DOTA-TATE: * p < 0.05, ** p < 0.01

Method:
Competition binding of SSTR2 ligands against [ 64 Cu]Cu-DOTA-TATE in MPC cell homogenatesCryo-preserved MPC cells (1×10 7 /mL) were resuspended in ice-cold RPMI 1640 medium supplemented with cOmplete™ EDTA-free proteinase inhibitor (Roche, Basel Switzerland) and homogenized using a Dounce homogenizer.A series of homogenates, each from 1×10 6 cells, were incubated with 2 nmol/L [ 64 Cu]Cu-DOTA-TATE (Am = 43.9MBq/nmol) in presence of additional competitor at increasing concentrations between 10 pmol/L and 5 µmol/L in a final volume of 0.5 mL for 1 hour at 37°C.Incubation was stopped by soaking the cell homogenates into Whatman GF/C collection filters (GE Healthcare, Chicago, IL, USA; presoaked in 0.3 % (v/v) polyethyleneimine for 90 min).Filters were washed with icecold Dulbecco's phosphate-buffered saline using a harvester (Brandel, Gaithersburg, MD, USA).Filterbound activity was measured using the γ counter Wizard (PerkinElmer), corrected for plastic-bound background, and normalized to bound activity without competitor.Inhibitory constants were calculated by non-linear regression analysis using the 'One-site Fit Ki' model implemented in Prism 10 with binding parameters for [ 64 Cu]Cu-DOTA-TATE constrained to Kd = 1.16 nmol/L and c = 2 nmol/L.Inhibitory constants of metal-free and metalated DOTA-TATE and NODAGA-TATE variantsIn the present study, competition binding against [ 64 Cu]Cu-DOTA-TATE showed similar inhibitory constants (Ki values between 0.76 and 1.78 nmol/L) for the non-radioactive metalated conjugates [ nat Cu]Cu-NODAGA-TATE, [ nat Cu]Cu-NODAGA-cLAB4-TATE and [ nat Lu]Lu-DOTA-TATE as well as for their corresponding metal-free precursors (Figure S 2).This indicates that the incorporated metal and the net charge of the metal complex have no influence on the binding of the different ligands versions towards SSTR2.This further suggests that the content of non-radioactive precursor in the radioligand preparations had no influence on affinity measurement in the present study.

Figure S 2 :
Figure S 2: Competitive binding of metal-free, lutetium, and copper-labeled TATE variants against [ 64 Cu]Cu-DOTA-TATE in MPC cell homogenates; competitors were investigated either metal-free or 'saturated' with indicated metal ions in naturally occurring isotope composition; (Ki) inhibition constants against [ 64 Cu]Cu-DOTA-TATE (Kd = 1.16 nmol/L, Am = 43.9MBq/nmol) at a molar concentration of 2 nmol/L; number of independent experiments (n) each performed in a triplicate; data presented as means with a confidence interval of 68 % Spheroid size was monitored and documented every 2−4 days after treatment start for up to 35 days using the microscope Axiovert 40 CFL(Carl Zeiss, Oberkochen, Germany).Bright field images were converted into ECAT files.Spheroids were identified and two-dimensional ROIs were drawn using a semiautomated contrast-based delineation workflow in Rover (ABX, Radeberg, Germany) with contrast thresholds between 1 and 180.Delineations were manually reviewed and adjusted in case of unfavorable contrast conditions, for example due to increased amounts of debris.Spheroids circular areas were extracted, and changes were analyzed in 32 spheroids per treatment condition.Fractions of controlled spheroids were determined for indicated time points.Progression-free control (PFC) was defined as < 1.2-fold change in spheroid circular area, overall control (OC) was defined as < 2.2-fold change in spheroid circular area.Median PFC and OC were calculated using the survival analysis tool implemented in Prism 10 (GraphPad Software).For calculation of half-maximal spheroid control doses (SCD50) at specific time points, controlled fractions were plotted against a series of increasing radioligand concentrations and analyzed by non-linear regression using the sigmoidal '[Agonist] vs. response -Variable slope (four parameters)' model implemented in Prism 10 (GraphPad Software).The 'Top' and 'Bottom' parameters were constrained to 1 and 0, respectively.The 'Hillslope' parameter was constrained to the best-fit value for [ 177 Lu]Lu-DOTA-TATE treatment.Uptake of copper-67-labeled NODAGA-TATE variants and growth inhibition in MPC tumor spheroidsThe activity concentration initially required for the copper-67-labeled NODAGA-TATE variants to provide adequate uptake and treatment efficacy was determined from growth inhibition of MPC tumor spheroids and compared to [ 177 Lu]Lu-DOTA-TATE (Figure S 3 A).Upon treatment with the highest tested activity concentration (400 kBq/mL), the effective uptake of the copper-67-labeled NODAGA-TATE variants in spheroids was 72 % or 49% lower compared to [ 177 Lu]Lu-DOTA-TATE, depending on their molar activity of 40 or 20 MBq/nmol, respectively (Figure S 3 B, Table S 2).Due to lower effective uptake, faster release, and shorter physical half-life of the radionuclide, the copper-67-labeled NODAGA-TATE variants required a four-to six-fold higher initial activity concentration to provide equivalent control of spheroid growth to [ 177 Lu]Lu-DOTA-TATE in vitro (Figure S 3 C−D, Table S 2).Further details on the analysis of growth-reducing effects on spheroids are provided in Figure S 4 A−C.This outcome is consistent with the lower binding capacities and faster release rates measured for the copper-64-labeled versions of the radioligands.Of note, the initial concentrations of copper-67-labeled NODAGA-TATE variants that provided half-maximal control of MPC spheroid growth are within the same range as has been recently reported for growth inhibition of genetically modified QGP1-SSTR2+ spheroids after treatment with the albumin-affine [ 67 Cu]Cu-DOTA-EB-TATE conjugate (200−1,000 kBq/mL) [50].The present results indicate that effective treatment with the copper-67-labeled NODAGA-TATE variants requires high initial radioligand concentrations and sufficiently long radioligand exposure in vivo to maximize effective initial uptake and to compensate the faster release from tumor cells.

Figure S 3 :
Figure S 3: Uptake and growth-inhibitory effects of lutetium-177-and copper-67-labeled TATE variants in MPC tumor spheroids; initial activity concentrations (IAC) 20−400 kBq/mL and initial activity doses (ID) 4−80 kBq/per microwell followed by radioligand removal; IAC corresponded to molar concentrations between 2−40 nmol/L (Am = 20 MBq/nmol) or 1−20 nmol/L (Am = 40 MBq/nmol); (A) Microscopic images of spheroid growth after radioligand treatment; in yellow: regions of interest to determine spheroid circular areas; (B) Specific binding and uptake in spheroids after treatment with radioligands at highest IAC; (C) Median progression-free control after treatment with radioligands at increasing IAC; (D) Dependence of progression-free control on the IAC of the radioligands 7 days after treatment start; progression-free control is defined as < 1.2-fold change in spheroid circular area; dotted lines indicate base-line values measured in spheroids without treatment; (Am) molar activity at treatment start; data presented as means ± standard error; significance of differences: * p < 0.05, ** p < 0.01, *** p < 0.001

Figure S 4 :
Figure S 4: Growth-inhibitory effects of lutetium-177-and copper-67-labeled TATE variants on MPC tumor spheroids; initial activity concentrations (IAC) 20−400 kBq/mL; initial activity doses (ID) 4−80 kBq/per microwell followed by radioligand removal from the culture medium; IAC corresponded to molar concentrations between 1−20 nmol/L (Am = 40 MBq/nmol) or 2−40 nmol/L (Am = 20 MBq/nmol); (A) Plate layout for exposure of spheroid arrays to increasing IAC of the radioligands; data presented as means ± standard error; (B) Changes in spheroid volume in response to treatments at increasing IAC; measured IAC in the incubation medium; horizontal lines indicate specific thresholds for survival analyses; (C) Treatment effects of the highest tested IAC on progression-free and overall control; progression-free control is defined as < 1.2-fold change in spheroid circular area compared to treatment start and overall control is defined as < 2.2-fold change in spheroid circular area [Xray], 113 , and 208 keV) and copper-67(93 and 185 keV).With each SPECT scan, a corresponding CT image was recorded and used for anatomical referencing and attenuation correction.Images were reconstructed using the Tera-Tomo™ three-dimensional (3D) algorithm with high dynamic range, a voxel size of 0.23 mm, and corrections for scatter and attenuation.Method: Spatial resolution measurements and activity calibrationFive-mL syringes filled with 2 mL of 0.154 mol/L NaCl(aq) containing 15 MBq of [ 177 Lu]LuCl3 or [ 67 Cu]CuCl2 served as resolution and activity phantoms.Spatial resolution of SPECT images was determined using the 'Resolution tool' implemented in Rover version 3.0.77h(ABX) as described elsewhere [83], with some modifications.In brief, a 20 mm square region-of-interest (ROI) was centered over the syringe contents and split into a series of ten axially stacked sub-ROIs.For each sub-ROI, the effective spatial resolution was determined from its radial voxel intensity profile by calculating the full width at half-maximum (FWHM) of the point spread function (PSF).The FWHM values of all sub-ROIs along the axial length of the syringe contents were averaged and expressed in mm.The activity values of voxel intensities in SPECT images were calibrated by determining the conversion factor between the activity of the syringe contents measured in an ISOMED 2010 dose calibrator (NUVIA Instruments, Dresden, Germany) and the corresponding total number of counts in a reconstructed SPECT image with a 50 mm axial FOV.

Figure S 5 :
Figure S 5: Spatial resolution and quantification accuracy in SPECT imaging of lutetium-177and copper-67; 5-mL-syringe activity phantoms containing activity concentration of 7.5 MBq/mL were measured using the Mediso nanoScan® SPECT/CT equipped with the APT56 aperture consisting of four multi-pinhole collimators; (A) Energy spectra, photon count rates, and 20 % energy windows of the recorded photopeaks for imaging of lutetium-177 (56, 113, and 208 keV) and copper-67 (93 and 185 keV); (B) Planar SPECT images of activity phantoms; dashed fields indicate ROIs used for image analysis; (C) Radial voxel intensity profiles showing the decrease in contrast at the edges of the activity phantoms along the transaxial radius of the ROI; (D) Effective spatial resolution along the axial length of the analyzed ROI, spatial resolution was determined as full width at half-maximum (FWHM) of the point spread function (PSF), means ± standard error; (E) Distribution of voxel intensities within the activity phantoms; the analyzed ROI included all voxels above the minimum intensity threshold of 30 %; (F) Statistics of the mean standardized uptake value (SUVmean) measured in the activity phantoms; median with 25 th and 75 th percentiles

Figure S 8 :
Figure S 8: Extraction of tissue-specific activity concentrations from SPECT/CT images of MPC allograft mice; positions of masks (red circles) and ROIs (yellow/white-highlighted regions) exemplified in orthogonal image slices of a SPECT/CT scan recorded 5 h after treatment with [ 67 Cu]Cu-NODAGA-cLAB4-TATE; intersections of red crosshairs indicate the positions of tissue-specific masks and ROIs; (L) left; (R) right, (CN) cranial, (CD) caudal

IX.
Figure S 11: Projected pharmacokinetics of lutetium-177-and copper-67-labeled TATE variants in female and male humans extrapolated from tissue-specific effective activity profiles in female MPC tumor allograft mice; animals received an initial activity dose of 50 MBq corresponding to a molar amount of 1.25 nmol (Am = 40 MBq/nmol); data presented normalized to initially injected activity dose without decay correction effects on subcutaneous MPC tumor allografts Treatments with [ 67 Cu]-NODAGA-TATE, [ 67 Cu]Cu-NODAGA-cLAB4-TATE, and [ 177 Lu]Lu-DOTA-TATE were associated with temporary inhibition of tumor growth resulting in treatment-specific improvements in survival (Figure S 12).Radioligands administered at a molar activity of 40 MBq/nmol provided the strongest growth-inhibiting effects including an initial reduction in tumor volume for up to 10 days in response to [ 67 Cu]Cu-NODAGA-cLAB4-TATE, and [ 177 Lu]Lu-DOTA-TATE.

Figure S 12 :
Figure S 12: Tumor growth and survival of MPC allograft mice treated with lutetium-177-and copper-67-labeled TATE variants; animals received an initial activity dose of 50 MBq corresponding to molar amounts of 1.25 nmol (Am = 40 MBq/nmol) or 2.5 nmol (Am = 20 MBq/nmol); (A) Fold-change in tumor volume after treatment; solid curves indicate the mean progression until ~8-fold change in tumor volume (> 60 % of animals available in each cohort); (B) Tumor start volumes in each cohort; (C) Overall survival of tumor-bearing mice is defined as < 10-fold change in tumor volume compared to treatment start; censored data points along horizontal lines indicate animals withdrawn from follow-up before reaching the overall survival criterion; (Am) molar activity at treatment start; data presented as means with standard error

Table S 1
: Animal cohorts for treatment with lutetium-177-and copper-67-labeled TATE variants b equivalent molar amount of radioligand administered at treatment start, respectively; (RN) radionuclide; (TV) tumor volume at treatment start; (n) number of animals per group and per independent experiment; data presented as means ± standard error III.

DOTA-/NODAGA-TATE variants and [ 177 Lu]Lu-DOTA-TATE.
antagonist [ 177 Lu]Lu-DOTA-JR11 compared to the agonist [ 177 Lu]Lu-DOTA-TATE, but not membrane preparations [44].However, radioligand-specific trends in binding capacities are reproducibly measured across a broad range of SSTR2 source material including intact MPC cells, cell homogenates, and tumor spheroids, thus excluding functional interference with the Bmax measurements in the present study.

Table S
2: Uptake of lutetium-177-and copper-67-labeled TATE variants in MPC tumor spheroids and inhibition of spheroid growth

67 Cu]Cu-NODAGA-TATE
(% ID) decay-corrected proportion of initial activity dose; data presented as means ± standard error

Table S 4
: Projected female and male human organ doses (mSv/MBq) calculated from extrapolated radioligand pharmacokinetics estimated for human females (60 kg) and males (73 kg) from quantitative analysis of SPECT images after treatment of female NMRI-nude mice with 50 MBq of the radioligand at a molar activity of 40 MBq/nmol; kinetic data input: heart contents, kidneys, liver, urinary bladder contents, total body remainder; data presented as means in mSv/MBq a