Influence of Residualizing Properties of the Radiolabel on Radionuclide Molecular Imaging of HER3 Using Affibody Molecules

Human epidermal growth factor receptor type 3 (HER3) is an emerging therapeutic target in several malignancies. To select potential responders to HER3-targeted therapy, radionuclide molecular imaging of HER3 expression using affibody molecules could be performed. Due to physiological expression of HER3 in normal organs, high imaging contrast remains challenging. Due to slow internalization of affibody molecules by cancer cells, we hypothesized that labeling (HE)3-ZHER3:08698-DOTAGA affibody molecule with non-residualizing [125I]-N-succinimidyl-4-iodobenzoate (PIB) label would improve the tumor-to-normal organs ratios compared to previously reported residualizing radiometal labels. The [125I]I-PIB-(HE)3-ZHER3:08698-DOTAGA was compared side-by-side with [111In]In-(HE)3-ZHER3:08698-DOTAGA. Both conjugates demonstrated specific high-affinity binding to HER3-expressing BxPC-3 and DU145 cancer cells. Biodistribution in mice bearing BxPC-3 xenografts at 4 and 24 h pi showed faster clearance of the [125I]I-PIB label compared to the indium-111 label from most tissues, except blood. This resulted in higher tumor-to-organ ratios in HER3-expressing organs for [125I]I-PIB-(HE)3-ZHER3:08698-DOTAGA at 4 h, providing the tumor-to-liver ratio of 2.4 ± 0.3. The tumor uptake of both conjugates was specific, however, it was lower for the [125I]I-PIB label. In conclusion, the use of non-residualizing [125I]I-PIB label for HER3-targeting affibody molecule provided higher tumor-to-liver ratio than the indium-111 label, however, further improvement in tumor uptake and retention is needed.


Introduction
Human epidermal growth factor receptor type 3 (HER3) is a transmembrane protein belonging to the human epidermal growth factor receptor (HER) family, which is involved in the regulation of cellular proliferation, motility, and apoptosis [1]. The intracellular kinase domain of HER3 is impaired and depends on heterodimerization with other members of the HER family for signal transduction [2]. Heterodimerization with HER3 appreciably enhances the signaling of other HER receptors, of which the HER2-HER3 and HER1 (EGFR)-HER3 heterodimers cause potent mitogenic of activity from these organs. This has been used to improve tumor-to-organ ratios for ESPs, e.g., tumor-to-liver [45] and tumor-to-kidney [46] ratios for affibody molecules, tumor-to-kidney ratio for ADAPTs [42], as well as tumor-to-kidney and tumor-to-liver ratios for DARPins [43,44]. It should be noted that this approach was applied for ESPs targeting HER2, which has a high level of receptor overexpression in vitro (e.g., 1.6 × 10 6 receptors/cell in SKOV-3 cells [47]) and in tumor models in vivo. In this study, we intended to test the hypothesis that the use of a non-residualizing radioiodine label would improve the tumor-to-liver ratio for the HER3-targeting Z HER3:08698 affibody molecule compared to a residualizing radiometal label.
To test this hypothesis, we used a (HE) 3 -Z HER3:08698 -DOTAGA affibody molecule, as this affibody molecule with the (HE) 3 -tag demonstrated the best tumor-to-liver ratio with an indium-111 label in the previous study [33]. Labeling it with indium-111 provided [ 111 In]In-(HE) 3 -Z HER3:08698 -DOTAGA, which was used as a control having residualizing label properties. To obtain a non-residualizing label, the same affibody molecule was used for indirect radioiodination using [ 125 I]-N-succinimidyl-4-iodobenzoate. This labeling approach was selected because the radiometabolites of [ 125 I]-4-iodobenzoate label are rapidly excreted and do no contribute to background [44,48]. Iodine-125 was used in this study for radioiodination. Due to long half-life and low-energy gamma emission, this nuclide is not suitable for imaging in humans. However, this nuclide is a convenient surrogate for 123 I and 124 I, which can be used for clinical SPECT and PET imaging, respectively. The DOTAGA chelator was not used for radiolabeling in this case but was utilized as a biodistribution modifier because an increased negative charge at the C-terminus reduces the hepatic uptake of Z HER3:08698 [33]. The (HE) 3 -tag at N-terminus was also used to reduce unspecific hepatic uptake [34].

In Vitro Studies
In vitro evaluation was performed using BxPC-3 (pancreatic cancer) and DU145 (prostate cancer) cells according to Wållberg and Orlova [50]. To demonstrate binding specificity of radiolabeled affibody molecules to HER3, the HER3 receptors were saturated with 1000-fold molar excess of a non-labeled HER3-targeting affibody molecule before addition of the radiolabeled conjugates. Blocking of the HER3 receptors resulted in a significant (p < 0.001, t-test) decrease of uptake of the radiolabeled molecules. This showed that both radiolabeled conjugates retained the HER3-specific binding after the labeling ( Figure 1). The uptake in BxPC-3 cells was higher than in DU145 cells. uptake of the radiolabeled molecules. This showed that both radiolabeled conjugates retained the HER3-specific binding after the labeling ( Figure 1). The uptake in BxPC-3 cells was higher than in DU145 cells. The binding kinetics of [ 125 I]I-PIB-(HE)3-ZHER3:08698-DOTAGA and [ 111 In]In-(HE)3-ZHER3:08698-DOTAGA to BxPC-3 cells were measured using LigandTracer [51]. The KD values for both conjugates were in the picomolar range (Table 2), with the indium-labeled conjugate having higher affinity than the radioiodinated one. The processing and internalization of radiolabeled affibody molecules by the HER3-expressing cells during continuous incubation is shown in Figure 2. For the radioiodinated conjugate, rapid association was followed by a plateau, while for the radiometal-labeled conjugate, an increase of cellassociated activity was observed over 24 h. The internalized fractions at 24 h were lower for affibody molecules labeled with non-residualizing iodine-125 label in comparison to residualizing indium-111 label. After 4 h of incubation, the internalized fraction for the indium-111 label was below 20% from total cell-associated activity in both cell lines. By 24 h, ca. 50% of cell-associated activity was internalized in BxPC-3 cells and ca. 30% was internalized in DU145 cells.  [51]. The K D values for both conjugates were in the picomolar range (Table 2), with the indium-labeled conjugate having higher affinity than the radioiodinated one. The processing and internalization of radiolabeled affibody molecules by the HER3-expressing cells during continuous incubation is shown in Figure 2. For the radioiodinated conjugate, rapid association was followed by a plateau, while for the radiometal-labeled conjugate, an increase of cell-associated activity was observed over 24 h. The internalized fractions at 24 h were lower for affibody molecules labeled with non-residualizing iodine-125 label in comparison to residualizing indium-111 label. After 4 h of incubation, the internalized fraction for the indium-111 label was below 20% from total cell-associated activity in both cell lines. By 24 h, ca. 50% of cell-associated activity was internalized in BxPC-3 cells and ca. 30% was internalized in DU145 cells.
Biodistribution of [ 125 I]I-PIB-(HE)3-ZHER3:08698-DOTAGA and [ 111 In]In-(HE)3-ZHER3:08698-DOTAGA was studied in Balb/c nu/nu mice bearing BxPC-3 xenografts at 4 and 24 h pi (Table 3). Both radiolabeled conjugates showed biodistribution characteristic for affibody molecules with rapid and predominantly renal clearance. However, there were a number of differences between the radioiodine-and the radiometal-labeled conjugates. The uptake of the radioiodine-labeled conjugate was generally several fold lower in normal organs and tissues, except blood. Especially prominent were the differences in kidney uptake, 291 ± 39%ID/g for the indium label vs. 2.7 ± 0.7%ID/g for the radioiodine label at 4 h pi, which further reduced to 0.15 ± 0.02%ID/g by 24 h. The tumor uptake was lower for the radioiodine label as well, 0.8 ± 0.1%ID/g vs. 2.4 ± 0.1%ID/g for the indium label at 4 h pi. The indium label also demonstrated better retention in the tumor than radioiodine at 24 h.
Due to lower retention in the blood and higher retention in the tumor, the radiometal label provided significantly (p < 0.05) higher tumor-to-blood, tumor-to-lung, and tumor-to-bone ratios than the radioiodine label at 4 h ( Table 4). Tumor-to-spleen and tumor-to-muscle ratios were comparable between the labels. On the other hand, the radioiodine label provided higher tumor-to-organ ratios in HER3-expressing organs (salivary gland, stomach, small intestine, and liver) at 4 h ( Table 4). By 24 h there was no marked increase in tumor-to-organ ratios for both labels, except the tumor-to-blood ratio for the 111 In-labeled conjugate.

In Vivo Studies
Both radiolabeled conjugates demonstrated specific binding in vivo to HER3-expressing BxPC-3 xenografts ( Figure 3). Uptake of activity in BxPC-3 xenografts was significantly (p < 0.0001) reduced when 72 µg of HER3 affibody molecule was injected compared to 2 µg.     (Table 3). Both radiolabeled conjugates showed biodistribution characteristic for affibody molecules with rapid and predominantly renal clearance. However, there were a number of differences between the radioiodine-and the radiometal-labeled conjugates. The uptake of the radioiodine-labeled conjugate was generally several fold lower in normal organs and tissues, except blood. Especially prominent were the differences in kidney uptake, 291 ± 39%ID/g for the indium label vs. 2.7 ± 0.7%ID/g for the radioiodine label at 4 h pi, which further reduced to 0.15 ± 0.02%ID/g by 24 h. The tumor uptake was lower for the radioiodine label as well, 0.8 ± 0.1%ID/g vs. 2.4 ± 0.1%ID/g for the indium label at 4 h pi. The indium label also demonstrated better retention in the tumor than radioiodine at 24 h. Due to lower retention in the blood and higher retention in the tumor, the radiometal label provided significantly (p < 0.05) higher tumor-to-blood, tumor-to-lung, and tumor-to-bone ratios than the radioiodine label at 4 h ( Table 4). Tumor-to-spleen and tumor-to-muscle ratios were comparable between the labels. On the other hand, the radioiodine label provided higher tumor-to-organ ratios in HER3-expressing organs (salivary gland, stomach, small intestine, and liver) at 4 h ( Table 4). By 24 h there was no marked increase in tumor-to-organ ratios for both labels, except the tumor-to-blood ratio for the 111 In-labeled conjugate.
Differences were significant (p < 0.05) between: a the uptake of (HE) 3

Discussion
The use of a novel class of imaging probes based on engineered scaffold proteins is approaching clinical practice [52]. Despite promising characteristics, only a few groups work with engineered scaffold proteins, and the available information concerning influence of different factors on their imaging properties is scarce. Particularly challenging is the development of probes for imaging of targets that are expressed not only in tumors but also in normal tissues, such as EGFR, IGF-1R, or HER3. However, the data suggest that an optimal molecular design and selection of an optimal labeling approach may noticeably enhance imaging contrast resulting in improved sensitivity. One of the substantial challenges in radionuclide molecular imaging of HER3 expression in tumors is to increase the tumor-to-liver ratio. Earlier, we investigated possibilities to minimize uptake of HER3 imaging probes in the liver. Our approach was based on the combination of dose adjustment (to minimize specific uptake) and optimization of molecular design, including selection of an optimal nuclide/chelator combination (to minimize unspecific uptake). This approach increased the tumorto-liver ratio by ca. 7-fold for EGFR-targeting affibody molecules from 0. 46 [41] in a BxPC-3 xenograft model. It has to be noted that although both EGFR and HER3 are expressed in normal tissues, including liver, the expression level of HER3 in tumors is

Discussion
The use of a novel class of imaging probes based on engineered scaffold proteins is approaching clinical practice [52]. Despite promising characteristics, only a few groups work with engineered scaffold proteins, and the available information concerning influence of different factors on their imaging properties is scarce. Particularly challenging is the development of probes for imaging of targets that are expressed not only in tumors but also in normal tissues, such as EGFR, IGF-1R, or HER3. However, the data suggest that an optimal molecular design and selection of an optimal labeling approach may noticeably enhance imaging contrast resulting in improved sensitivity. One of the substantial challenges in radionuclide molecular imaging of HER3 expression in tumors is to increase the tumor-to-liver ratio. Earlier, we investigated possibilities to minimize uptake of HER3 imaging probes in the liver. Our approach was based on the combination of dose adjustment (to minimize specific uptake) and optimization of molecular design, including selection of an optimal nuclide/chelator combination (to minimize unspecific uptake). This approach increased the tumor-to-liver ratio by ca. 7-fold for EGFR-targeting affibody molecules from 0. 46 [41] in a BxPC-3 xenograft model. It has to be noted that although both EGFR and HER3 are expressed in normal tissues, including liver, the expression level of HER3 in tumors is appreciably lower than the level of EGFR. In this study, we evaluated an alternative strategy aiming towards decreasing the retention of the radionuclide in the liver rather than reducing its uptake. To achieve this, we included two modifications in the molecular design of a new tracer, that were previously shown to decrease hepatic uptake, i.e., (HE) 3 -tag at the N-terminus and DOTAGA chelator at the C-terminus.
The use of [ 125 I]N-succinimidyl-4-iodobenzoate resulted in a stable labeling of (HE) 3 -Z HER3:08698 -DOTAGA (Table 1). In vitro binding of [ 125 I]I-PIB-(HE) 3 -Z HER3:08698 -DOTAGA to HER3-expressing cell lines was specific as demonstrated in a saturation experiment (Figure 1). The affinity of the radioiodinated variant was lower compared to the 111 In-labeled counterpart ( Table 2), but remained high, 98 ± 12 pM, which was considered as a precondition for successful in vivo targeting. The results of in vitro evaluation of cellular processing were somewhat surprising, as the internalization of [ 111 In]In-(HE) 3 -Z HER3:08698 -DOTAGA was somewhat faster (54% at 24 h for BxPC-3 cells) compared with the earlier reported internalization of [ 111 In]In-Z HER3:08698 -DOTAGA without (HE) 3 -tag (around 30% at 24 h for BxPC-3 cells) [33]. However, only 18% of [ 111 In]In-(HE) 3 -Z HER3:08698 -DOTAGA was internalized at 4 h, which indicated that the use of a non-residualizing label would not be associated with a major loss of cell-associated activity within a few hours after injection. Although the general pattern of cellular processing of [ 111 In]In-(HE) 3 (Table 3). Already at 4 h pi the activity of 125 I in the majority of normal tissues was substantially lower compared to 111 In. The most striking difference was the more than 100-fold reduction of renal activity. The reduction of activity in other HER3-expressing tissues (salivary gland, liver, stomach, small intestines) was 6-10-fold for the non-residualizing [ 125 I]I-PIB label. Decrease of the liver uptake was approximately 10-fold. The level of blood-born activity was higher for the radioiodinated variant. This is most likely caused by released radioiodine metabolites from kidneys and liver while the radiometabolites of 111 In-label were trapped inside tissues where [ 111 In]In-(HE) 3 -Z HER3:08698 -DOTAGA was taken up. Earlier, we have shown for another scaffold protein with high renal uptake, ADAPT, that the rapid decrease of activity in kidneys for a non-residualizing radioiodine label is associated with higher percentage of low-molecular-weight fraction in blood in comparison with a radiometal label [42]. The tumor uptake was also lower for [ 125 I]I-PIB-(HE) 3 -Z HER3:08698 -DOTAGA, but the decrease was 3.5-fold, i.e., smaller compared to decrease for normal tissues. Thus, the ratios of uptake in tumors to uptake in majority of tissues were higher for [ 125 I]I-PIB-(HE) 3 -Z HER3:08698 -DOTAGA compared to [ 111 In]In-(HE) 3 -Z HER3:08698 -DOTAGA at 4 h pi ( Table 4). The exceptions were tumor-to-blood and tumor-to-bone ratios. A SPECT/CT image (Figure 4) confirmed the biodistribution data. At 24 h after injection, there was a substantial reduction of tumor uptake of [ 125 I]I-PIB-(HE) 3 -Z HER3:08698 -DOTAGA compared to 4 h, which reflects the increasing internalization of the tracer and leakage of radiometabolites. It has to be noted that the increase of tumor-to-organ ratios for [ 125 I]I-PIB-(HE) 3 -Z HER3:08698 -DOTAGA at 4 h pi was obtained at the cost of noticeable decrease of tumor uptake. The decrease in tumor uptake was more substantial than it could be expected from in vitro cellular processing data. As cells in vivo are influenced by a variety of signaling substances, such influence is difficult to mimic in vitro (if possible at all). The increased internalization rate of HER3 or HER3/[ 125 I]I-PIB-(HE) 3 -Z HER3:08698 -DOTAGA adduct might be a consequence of such influence. This suggests that in vitro modeling would not be quite predictive for in vivo situation. Nevertheless, the tumor-to-liver ratio of 2.4 ± 0.3 was achieved for [ 125 I]I-PIB-(HE) 3 -Z HER3:08698 -DOTAGA, which is one of the best for anti-HER3 affibody molecules. Overall, the results of this study suggest that the hypothesis that tumor-to-liver and other tumor-to-organ ratios can be improved by the use of a non-residualizing label is correct.
While this study demonstrated feasibility of the use of non-residualizing labels for enhancing the tumor-to-liver ratio of anti-HER3 affibody molecule, the low tumor uptake is of concern. Further strategies for development of this approach could include modulation of radiometabolite retention and increase of affinity. [ 125 I]I-para-iodobenzoate is only one of the potential pendant groups providing a non-residualizing radioiodine label. Alternatively, labeled iodo-hydroxyphenylethyl-maleimide (I-HPEM) [55,56] or 4-iodophenethylmaleimide (I-PEM) [57] can be used for indirect radioiodination of affibody molecules. These alternative pendant groups have different rates of excretion from normal tissues compared to PIB, which opens an opportunity for finding the most suitable variant. Non-residualizing properties are also typical for radiofluorine labels [58][59][60]. Importantly, all these labels are site-specific, and their use might help to avoid undesirable lysine modifications close to the binding site of the affibody molecule and might increase the binding affinity to HER3. This might also increase the tumor retention of activity. Furthermore, although (HE) 3 -tag reduced non-specific hepatic uptake of affibody molecules, the presence of (HE) 3 -tag accelerated the internalization rates. Thus, the use of an affibody without (HE) 3 -tag should be evaluated in combination with non-residualizing labels.
The use of murine model in this feasibility study was based on an assumption that rates of internalization of affibody molecules by human and murine hepatocytes are similar. However, the rates might be different in rodents and humans. A dual-label microdosing clinical study would be necessary for confirmation of translational potential of our findings.
Although outside the scope of this study, it is worth mentioning that [ 111 In]In-(HE) 3 -Z HER3:08698 -DOTAGA demonstrated an overall lower uptake and retention in normal organs (ca. eight times lower in blood, two times lower in lungs and liver at 4 h pi) in comparison with the previously reported [ 111 In]In-Z HER3:08698 -DOTAGA HER3 affibody molecule without the (HE) 3 -tag [33]. Even though the tumor uptake of the (HE) 3 -tagged variant was lower (2.4 ± 0.1%ID/g vs. 3.2 ± 0.1%ID/g), the tumor-to-organ ratios were higher. Moreover, [ 111 In]In-(HE) 3 -Z HER3:08698 -DOTAGA provided higher tumor-to-organ ratios than any 111 In-labeled Z HER3:08698 variant tested before [33]. For example, the tumor-to-blood ratio of 88 ± 9 (24 h pi) was more than four-fold higher than any value obtained previously for any Z HER3:08698 derivative. These data emphasize the strength of molecular design optimization.

General Materials and Instruments
Indium chloride [ 111 In]InCl 3 was purchased from Mallinckrodt Pharmaceuticals (Staines-upon-Thames, UK). Sodium iodide [ 125 I]NaI was purchased from Perkin Elmer Sverige AB (Hägersten, Sweden). Instant thin-layer chromatography (iTLC) analysis was performed using iTLC silica gel strips (Varian, Lake Forest, CA, USA). The activity distribution was measured using a Cyclone storage phosphor system (Packard) and analyzed by OptiQuant image analysis software (Perkin Elmer, Waltham, MA, USA). Purification was performed using NAP-5 columns (GE Healthcare, Little Chalfont, United Kingdom) pre-equilibrated with 1% BSA in PBS and eluted with PBS. Activity was measured using an automated gamma-spectrometer with a NaI(TI) detector (1480 Wizard, Wallac, Finland). BxPC-3 and DU145 cells were purchased from the American Type Culture Collection (ATCC) and were cultured in RPMI medium supplemented with 10% fetal bovine serum (FBS), 2 mM L-glutamine, 100 IU/mL penicillin, and 100 µg/mL streptomycin in a humidified incubator with 5% CO 2 at 37 • C, unless stated otherwise.

Binding Specificity and Cellular Processing Assays
In vitro studies were performed using HER3-expressing cancer cell lines BxPC-3 (12 ± 2 × 10 3 receptors/cell) [31] and DU145 (estimated as 60% of the expression level in BxPC-3 cells). Cells were seeded in 3 cm petri dishes (ca. 10 6 cells per dish), a set of three dishes was used for each group.
Binding specificity to HER3 was evaluated as described previously [50]. To saturate HER3 receptors, 1000-fold excess of non-labeled anti-HER3 affibody molecule (50 nM) was added to one group of cells, media was only added to the second group. After 15 min incubation at room temperature, radiolabeled [ 125 I]I-PIB-(HE) 3 -Z HER3:08698 -DOTAGA or [ 111 In]In-(HE) 3 -Z HER3:08698 -DOTAGA was added to both groups at 0.05 nM final concentration. After incubation for 1 h at 37 • C, the cell medium was collected, cells were washed with 1 mL of medium, and 1 mL of 1 M NaOH was added to lyse the cells. After 30 min of incubation, the cell lysate was collected. The radioactivity in each fraction was measured to calculate the percent of cell-bound radioactivity.
Cellular retention and processing of radiolabeled affibody molecules was studied during continuous incubation by an acid-wash method [50]. Radiolabeled affibody molecules (0.1 nM) were added to cells and incubated at 37 • C in a humidified incubator for 1, 2, 4, 8, and 24 h. At the selected time point, the media was collected from one set of dishes and cells were washed once with serum-free media (1 mL). To collect the membrane-bound fraction, the cells were treated with 0.2 M glycine buffer containing 4 M urea, pH 2.0 (1 mL) on ice for 5 min, the buffer was collected, and the cells were washed once with the same buffer (1 mL). To collect the internalized fraction, the cells were treated with 1 M NaOH (1 mL) for 30 min, the cells were collected and additionally washed with 1 mL. The activity in every fraction was measured and the percentage of cell-associated activity was calculated. Every dataset in Figure 2 represents the processing of one radiolabeled conjugate in one cell line. The maximum value of cell-associated activity in each dataset (at 8 or 24 h) was taken as 100% and the other dataset values were normalized to it.

Affinity Measurements Using LigandTracer
The binding kinetics of radiolabeled affibody molecules to living BxPC-3 cells was measured using LigandTracer and evaluated using the TraceDrawer Software (all from Ridgeview Instruments, Vänge, Sweden) as described earlier [51]. Increasing concentrations of radiolabeled affibody molecules (1 and 5 nM) were added to cells, followed by the change of media and measurements of retention in the dissociation phase. Kinetics was recorded at room temperature and dissociation constants were calculated based on association and dissociation rates.

Animal Studies
The animal experiments were planned and performed according to national legislation on protection of laboratory animals. The animal studies were approved by the local ethics committee for animal research in Uppsala, Sweden (ethical permission C5/16 from 26-02-2016).
For implantation of tumors, 5 × 10 6 of BxPC-3 cells in 100 µL of media were subcutaneously injected in the right hind leg of female Balb/c nu/nu mice. The experiments were performed three weeks after implantation. The average animal weight was 17 ± 1 g. The average tumor weight was 0. To demonstrate the HER3-mediated uptake, the protein dose was increased to 72 µg using non-labeled anti-HER3 affibody molecule in one group of mice. At 4 and 24 h post-injection (pi) mice were anesthetized by an intraperitoneal injection of ketamine and xylazine solution and sacrificed by heart puncture. The dose of ketamine was 250 mg/kg, and the dose of xylazine was 25 mg/kg. The organs and tissues were collected, weighed, and the activity was measured using an automated γ-spectrometer. The percentage of injected dose per gram of sample (%ID/g) was calculated. Data for the intestines with contents and carcass were calculated as %ID per whole sample.
Whole body SPECT/CT scans of mice bearing BxPC-3 tumors and injected with [ 125 I]I-PIB-(HE) 3 -Z HER3:08698 -DOTAGA (2 µg, 1.9 MBq) or with [ 111 In]In-(HE) 3 -Z HER3:08698 -DOTAGA (2 µg, 1.5 MBq) were performed using nanoScan SPECT/CT (Mediso Medical Imaging Systems, Hungary). Imaging at 4 and 24 h pi was performed after mice were sacrificed by CO 2 and urinary bladders were removed. The acquisition time was 20 min. In the case of 125 I, energy window between 26 and 31 keV was used. For 111 In, gamma-peaks of 245 and 171 keV (window width of 20%) were used for acquisition. CT scans were acquired using the following parameters: X-ray energy peak of 50 keV; 670 µA; 480 projections; and 5.26 min acquisition time. SPECT raw data were reconstructed using Tera-Tomo™ 3D SPECT reconstruction technology (version 3.00.020.000; Mediso Medical Imaging Systems Ltd., Budapest, Hungary): normal dynamic range; 30 iterations; and one subset. CT data were reconstructed using Filter Back Projection in Nucline 2.03 Software (Mediso Medical Imaging Systems Ltd., Budapest, Hungary). SPECT and CT files were fused using Nucline 2.03 Software (Mediso Medical Imaging Systems Ltd., Budapest, Hungary) and are presented as maximum intensity projections in the RGB color scale.

Statistical Analysis of the Data
Statistical analysis was performed using GraphPad Prism (version 7.02; GraphPad Software, Inc., La Jolla, CA, USA). p < 0.05 was considered a statistically significant difference. The in vitro data were analyzed using an unpaired two-tailed t-test. A paired two-tailed t-test was applied for the analysis of biodistribution data from the dual-label study to find significant differences.

Conclusions
In conclusion, this study demonstrated the feasibility of increasing tumor-to-liver contrast for HER3-targeting affibody molecules by the use of a non-residualizing label. However, further studies are required to provide better uptake in tumors.