The GEM-handle as convenient labeling strategy for bimodal single-domain antibody-based tracers carrying 99mTc and a near-infrared fluorescent dye for intra-operative decision-making

Intra-operative fluorescence imaging has demonstrated its ability to improve tumor lesion identification. However, the limited tissue penetration of the fluorescent signals hinders the detection of deep-lying or occult lesions. Integrating fluorescence imaging with SPECT and/or intra-operative gamma-probing synergistically combines the deep tissue penetration of gamma rays for tumor localization with the precision of fluorescence imaging for precise tumor resection. In this study, we detail the use of a genetically encoded multifunctional handle, henceforth referred to as a GEM-handle, for the development of fluorescent/radioactive bimodal single-domain antibody (sdAb)-based tracers. A sdAb that targets the urokinase plasminogen activator receptor (uPAR) was engineered to carry a GEM-handle containing a carboxy-terminal hexahistidine-tag and cysteine-tag. A two-step labeling strategy was optimized and applied to site-specifically label IRDye800CW and 99mTc to the sdAb. Bimodal labeling of the sdAbs proved straightforward and successful. 99mTc activity was however restricted to 18.5 MBq per nmol fluorescently-labeled sdAb to prevent radiobleaching of IRDye800CW without impeding SPECT/CT imaging. Subsequently, the in vivo biodistribution and tumor-targeting capacity of the bimodal tracer were evaluated in uPAR-positive tumor-bearing mice using SPECT/CT and fluorescence imaging. The bimodal sdAb showed expected renal background signals due to tracer clearance, along with slightly elevated non-specific liver signals. Four hours post-injection, both SPECT/CT and fluorescent images achieved satisfactory tumor uptake and contrast, with significantly higher values observed for the anti-uPAR bimodal sdAb compared to a control non-targeting sdAb. In conclusion, the GEM-handle is a convenient method for designing and producing bimodal sdAb-based tracers with adequate in vivo characteristics.


Introduction
Cancer surgery remains central in the curative treatment of solid tumors.Yet, specific and accurate intra-operative tumor lesion identification and margin delineation remain challenging.Unnoticed lesions and incorrect delineations can lead to tumor recurrence or iatrogenic tissue damage.Positive resection margins, occurring in 5-35% of operated patients, significantly increase the risk of local cancer relapse, impact the aggressiveness of postsurgical chemotherapy, and decrease overall and disease-specific survival rates (1).Over the last decade, fluorescence-guided imaging has emerged as the front-runner in intra-operative tumor imaging.It has demonstrated its efficacy to provide real-time visualization of tumor lesions without disrupting the surgical field or workflow (2)(3)(4)(5)(6).However, the limited tissue penetration of even near-infrared fluorescent signals remains a constraint for identifying occult tumor lesions beyond a few millimeters of depth.This can be overcome by combining fluorescence detection with pre-operative nuclear imaging and/or intra-operative gamma-probing.This combination offers an optimal complementary relationship between the high sensitivity of gamma-ray detection for the localization of hidden tumor lesions and the visual cues provided by fluorescent signals for precise tumor resection (7,8).In recent years, non-targeted ICG-[ 99m Tc]Tc-nanocolloids have successfully improved sentinel lymph node mapping through their bimodal nature compared to the separate modalities; thus advocating for a combined use of both modalities (9)(10)(11)(12).
To further implement bimodal cancer imaging in clinical practice, bimodal tracers based on tumor marker-directed targeting moieties are essential to adequately identify and resect lesions.For their design, both labels are preferentially attached onto the same targeting moiety to ensure consistent pharmacokinetic behavior.The simplest approach involves random conjugation of both labels separately onto the targeting moiety (13)(14)(15)(16).However, this results in heterogeneous end products and may negatively impact the functionality or pharmacokinetic behavior of small targeting moieties (17,18).A more controlled approach employs trivalent platforms that combine a fluorophore, a radiolabeling site and an attachment site on a single scaffold (19)(20)(21)(22)(23)(24)(25)(26)(27)(28).This approach ensures precise and consistent spatial positioning of the labels relative to each other.The trivalent platforms can then be attached to the targeting moiety either randomly or in a sitespecific manner.However, the production of trivalent platforms is labor-intensive as it involves a multi-step synthesis.More recently, platforms have been explored that utilize the fluorophore itself as the linking scaffold between the targeting moiety and the chelator, reducing the overall size of the construct (28)(29)(30).
In this study, we introduce an alternative and more convenient approach for designing bimodal tracers, namely the Genetically Encoded Multifunctional-handle (GEM-handle).The amino acid sequence of peptide-or protein-based targeting moieties can be readily engineered to encode additional amino acid motives, which enable labeling through various site-directed chemistries.Since these sites are inherent to the overall structure of the molecule, they ensure a consistent positioning of the labels on the targeting moiety.Our GEM-handle comprises two labeling sites, a hexahistidine-tag and a cysteine-tag, separated by a 14-amino acid spacer (Figure 1).The cysteine-tag allows for labeling with any maleimide-functionalized near-infrared fluorophore of interest, while the hexahistidine-tag exclusively permits labeling with technetium-99m ( 99m Tc) (31).Notably, 99m Tc is a low-energy gamma-ray emitter routinely employed in clinical practice for both pre-operative SPECT/CT imaging and intra-operative gamma-probing.This makes this isotope well-suited for the intended purpose of bimodal tracers.
Our GEM-handle leverages the carboxy-terminal tags we routinely incorporate for site-specific labeling on single-domain antibodies (sdAbs).sdAbs are ideal small antibody fragments derived from camelid heavy-chain antibodies for the development of targeted bimodal tracers due to their distinct advantages: Schematic representation of our GEM-handle engineered into a sdAb and its potential applications.remarkable target specificity, compact size and ease of engineering.Unlike antibodies, which remain in circulation for days and tend to accumulate non-specifically, sdAbs enable high-contrast imaging within hours after injection and with minimal background because of their rapid renal clearance.Moreover, in comparison with peptides, sdAbs serve as a platform technology that allows straightforward generation of sdAbs targeting almost any biomarker of interest, along with predictable pharmacokinetics.Previous preclinical and clinical investigations involving sdAbbased tracers for fluorescent, nuclear or bimodal imaging have shown the potential of these targeting moieties (3,27,(32)(33)(34)(35)(36)(37)(38)(39).In this paper, we report on our GEM-handle labeling approach for achieving bimodal labeling of sdAb-based tracers, focusing on a sdAb previously designed to target the urokinase plasminogen activator receptor (uPAR) (40).Furthermore, we evaluate its in vivo biodistribution and effectiveness in targeting tumors in a subcutaneous mouse model.
2 Materials and methods

GEM-handle modified sdAb production
The anti-human/canine uPAR sdAb uPAR15 was previously generated and validated preclinically by Mateusiak et al. (40).uPAR15 and the control sdAb R3b23 were cloned into a pHEN25 plasmid coding for a carboxy-terminal tail encoding a hexahistidine-tag and a cysteine-tag separated by a rigid 14 amino acid linker based on the hinge region of human IgA1 (Figure 1) (39,41).The sdAb-GEM constructs were produced and purified according to a previously described method (39,41).Briefly, sdAb-GEM constructs were expressed in E. coli and purified from periplasmic extracts using immobilized metal affinity chromatography followed by size exclusion chromatography (SEC).After purification, uPAR15-GEM and R3b23-GEM presented in monomeric form (with the cysteine-tag linked to glutathione) and dimeric form (two sdAbs linked by their cysteine-tags).Both forms were collected as separate fractions.

Two-step bimodal labeling of the sdAb-GEM 2.2.1 Reduction of the cysteine-tag and fluorescence labeling
As a first step in the bimodal labeling procedure, maleimidecysteine chemistry was employed to label uPAR15-GEM and R3b23-GEM site-specifically with IRDye800CW (Li-COR Biosciences; Nebraska, USA).SdAb-GEM (3 mg) was incubated with a 90-or 180-fold molar excess of 2-mercaptoethylamine (Thermo Fisher Scientific; Massachusetts, USA) for dimer or monomer fractions, respectively, and EDTA (5 mM, Sigma-Aldrich; Missouri, USA) in PBS (2.5 mL, pH 7.4; Thermo Fisher Scientific) at 37°C for 90 min.A PD-10 desalting column (Cytiva; Hoegaarden, Belgium) was equilibrated with ammonium acetate buffer (25 mL, 0.2 M, pH 6.0; Sigma-Aldrich) and subsequently, a buffer exchange was performed by applying the reduced sdAb-GEM and eluting with ammonium acetate (3.5 mL 0.2 M, pH 6.0).The collected sdAb-GEM was transferred to a Vivaspin column (Vivaspin 25,000 MWCO HY; Sartorius; Göttingen, Germany) and concentrated to a volume of 0.5 mL.Then, the reduced sdAb-GEM was incubated with a 5-fold molar excess of IRDye800CW-maleimide (20 mg/mL in DMSO) and EDTA (5 mM, 1 mL in 0.2 M ammonium acetate, pH 6.0) in a lightprotected vial at 37°C for 120 min.Purification of the sdAb-GEM-IRDye800CW from the excess IRDye800CW was performed via SEC on a Superdex Increase 75 10/300 GL (Cytiva) with PBS (pH 7.4) as a running buffer (0.8 mL/min).The sdAb-GEM and dye concentration of the IRDye800CW-labeled compounds was calculated through absorption measurement at 280 nm for the sdAb and at 774 nm for IRDye800CW using spectrophotometry (Nanodrop 2000; Thermo Fisher Scientific).To attribute for the absorption of IRDye800CW at 280 nm, the value measured at 280 nm was corrected by 3% of the value at 774 nm before calculation.The degree of labeling of the constructs was determined as the concentration of IRDye800CW to sdAb-GEM (42).
Quality control (QC) of each sdAb-GEM-IRDye800CW was performed via SEC on a Superdex Increase 75 10/300 GL (Cytiva) with PBS (pH 7.4) as a running buffer (0.8 mL/min).Detection was performed through absorption at 280 nm and 774 nm.The purity of the sdAb-GEM-IRDye800CW was considered adequate at 95% or higher, determined as the percentual area-under-the-curve (% AUC) of the sdAb-GEM-IRDye800CW peak on the QC SEC profile.
The post-radiolabeling fluorescent signal was compared to the signal of the positive control using the Odyssey scanner 9120 (Li-COR).Additionally, to ensure the temperature and kit buffer did not negatively affect the fluorescent signal of the compounds, radiolabeling was performed at 37 and 21°C compared to 50°C, and the sdAb-GEM-IRDye800CW constructs were incubated in a decayed kit under the above-described conditions.

In vivo biodistribution and tumor targeting of bimodal sdAb
All animal studies were performed according to the European Directive 2010/63/EU and received approval from the Ethical Commission for Animal Experimentation of the Vrije Universiteit Brussel (project nr.21-272-13).Female Crl : NU-Foxn1nu mice were purchased from Charles River at 6 weeks old (18 -25 g).The mice were group housed in individually ventilated cages at 19 to 24°C and 40 to 60% humidity with 4 mice per cage.A light/dark cycle of 14/ 10 h was implemented.Low-fluorescence pellet food (Teklad 2016, Basis Global Technologies; Illinois, USA) and water were provided ad libitum.After tumor inoculation and growth to 200 -500 mm 3 , the mice were randomly allocated to the experimental and control groups (4 mice per group) by a blinded laboratory technician.Upon inclusion in the experiment, starting from the tumor inoculation, all mice were inspected daily by assessment of behavior, appearance, and tumor growth.Tumors with a size above 100 mm 3 were measured every two to three days.The humane endpoints applied in this study were 1) a tumor size above 1500 mm 3 or a tumor ulceration above 10% of the tumor volume, 2) a body condition score of 1, and 3) a physical appearance or abnormal behavior indicative of pain or sickness.All mice in the study were killed through cervical dislocation.Injections, imaging, and killing of the mice were carried out under isoflurane anesthesia (5% induction, 2% maintenance, 1.0 L/min O 2 ).

Longitudinal assessment of the in vivo biodistribution and tumor targeting capacity of anti-uPAR bimodal sdAb
HT-29 cells (ATCC; Virginia, USA) were cultured in McCoy's medium supplemented with fetal bovine serum, glutamine, and penicillin/streptavidin at 37°C and 5% CO 2 .Female Crl : NU-Foxn1nu mice (n = 4, N = 16) were subcutaneously inoculated with 2 x 10 6 uPAR-positive HT-29 cells in the right flank and allowed to grow till 200 -500 mm 3 .2 nmol uPAR15-GEM-[ 99m Tc]Tc(CO) 3 -IRDye800CW or R3b23-GEM-[ 99m Tc]Tc(CO) 3 -IRDye800CW (23 ± 2 MBq) was intravenously injected via the tail vein.SPECT/CT imaging was performed 1, 4, and 8 h post-injection (p.i.) and fluorescence imaging was performed 1, 4, 8, and 24 h p.i.For each bimodal tracer, a group of mice was killed at 1 and 24 h p.i. and relevant organs and tissues were collected for ex vivo analysis through immediate fluorescence imaging and 99m Tc gammacounting post-dissection.Ex vivo radioactive uptake values for the tumor and relevant organs were decay-corrected and are described as %ID/g.

Imaging protocol and analysis
SPECT/CT imaging was performed for 25 min using the Vector + microSPECT/CT system from MILabs (Houten, Netherlands).The system was fitted with a rat/mouse 75 pinhole collimator (1.5 mm).Spiral mode SPECT scans were performed over 6 bed positions (200 s/position).CT scans were performed immediately after the SPECT scans (60 kV, pixel size = 80 µm).SPECT scans were reconstructed with the SPECT-Rec software (MILabs) after acquisition (subsets = 2, iterations = 4, voxel size = 0.4 mm) and paired with the corresponding CT scans.Further image analysis was handled in the Amide and OsiriX software.3D regions-of-interest (ROIs) were allocated on the SPECT images for the tumor and the %ID/cc in the ROIs was determined.
In vivo and ex vivo fluorescence imaging was performed using the FluoBeam800 (Fluoptics; Grenoble, France) in a dark room.White light images were obtained under normal room light.Acquisition was performed on the raw data setting with exposure times of 25 -300 ms.Fluorescent images were analyzed in ImageJ (Fiji).2D ROIs were drawn onto the white light images around the tumor, the muscle of the opposing hind leg or the relevant organs.The ROIs were transferred onto the fluorescent images for mean fluorescent intensity (MFI) determination.In vivo tumor-tobackground ratios (TBRs) were calculated as the ratio of the tumor MFI to the contralateral muscle MFI.

Statistical analysis
In vivo fluorescence TBRs of the experimental and control groups were compared using an unpaired student t-test per imaging modality.In vivo and ex vivo overall uptake values and absolute MFI values were compared using an unpaired student Ttest.All statistical analyses were performed using GraphPad Prism (version 8.4.3., GraphPad Software; California, USA) with p values: *p<0.05,**p<0.01,***p<0.001,and ****p<0.0001.All data is presented as mean ± SD and the graphical representations of the data were made with GraphPad Prism 8.4.3.

Fluorescence labeling of sdAb-GEM
For this study, the sdAb-GEM were labeled with IRDye800CW and [[ 99m Tc]Tc(H 2 O) 3 (CO) 3 ] + via a two-step labeling strategy utilizing respectively the cysteine-tag and hexahistidine-tag embedded in the GEM-handle.Firstly, uPAR15-GEM and R3b23-GEM were successfully labeled with IRDye800CW as indicated by the two coinciding peaks at 280 nm (sdAb-GEM) and 774 nm (IRDye800CW) on the SEC QC profiles (Figure 2A; Supplementary Figure 1A).Based on the %AUC, the purity of uPAR15-GEM-IRDye800CW and R3b23-GEM-IRDye800CW was respectively 97.8% and >99.9%.The degree of labeling of the respective sdAb-GEM-IRDye800CW constructs was 0.9 and 0.8.A minor fraction of unconjugated sdAb-GEM remained in the mixture, which could not be further purified through SEC due to the minor molecular weight difference.

Radiolabeling optimization of sdAb-GEM-IRDye800CW
In the second step, the sdAb-GEM-IRDye800CW were labeled with 99m Tc via their His-tag, achieving a radiochemical purity of > 87.0% as determined by iTLC analysis.For subsequent in vivo applications, the compounds were further purified to achieve a radiochemical purity exceeding 99% (Figure 2B; Supplementary Figure 1B).Nevertheless, it was noted that the fluorescent signal of the sdAb-GEM-IRDye800CW was influenced by the 99m Tclabeling, with the extent of the impact being dependent on the added amount of radioactivity (priorly, it had been confirmed that the incubation temperature and the composition of the buffer in the lyophilized Isolink kit for 99m Tc labeling did not impact the fluorescence of sdAb-GEM-IRDye800CW (data not shown)).The fluorescent signal of 2 nmol of uPAR15-GEM-[ 99m Tc]Tc(CO) 3 -  IRDye800CW remained similar to the reference uPAR15-GEM-IRDye800CW after radiolabeling with 18.5 MBq/nmol in 250 µL and remained stable for at least 6h after incubation at room temperature.At 37 MBq/nmol the fluorescent signal was preserved until 1h post-radiolabeling; however, a gradual decrease in signal was observed at later time points (Figures 2C, D).For conditions above 37 MBq/nmol, an almost complete loss of fluorescent signal was observed immediately after radiolabeling (Figures 2C, D).This corroborates with the SEC profiles of the uPAR15-GEM-[ 99m Tc]Tc(CO) 3 -IRDye800CW samples showing a decline in absorption at 774 nm similar to the decline in fluorescent signal after radiolabeling (Supplementary Figure 2).Interestingly, the radiolabeling could be upscaled by increasing the mass of sdAb-GEM-IRDye800CW and the reaction volume.As such, incubating 12 nmol of sdAb-GEM-IRDye800CW with 222 MBq (18.5 MBq/ nmol) in 500 µL proved possible with preservation of the fluorescent signal of the bimodal tracer (Figure 2B; Supplementary Figure 1B).This is necessary for in vivo studies.

Longitudinal in vivo biodistribution and tumor targeting capacity of anti-uPAR bimodal sdAbs
Finally, the biodistribution and tumor-targeting capacity of uPAR15-GEM-[ 99m Tc]Tc(CO) 3 -IRDye800CW were evaluated in vivo in HT-29 tumor-bearing mice and compared to the nontargeting R3b23-based compound.SPECT/CT and fluorescent images indicate that both uPAR15-and R3b23-GEM-[ 99m Tc]Tc (CO) 3 -IRDye800CW were rapidly eliminated from the circulation via the kidneys (188 ± 24%ID/g at 1h p.i.) and, to a much lesser extent, the liver (1.9 ± 0.2%ID/g at 1h p.i.).The compounds showed significant excretion through urine, yet a substantial portion remained in the kidneys for at least 24 h p.i. (150 ± 18%ID/g) (Figures 4A, B; Supplementary Figure 3).Minimal uptake was seen in other organs.These observations were confirmed by ex vivo analysis at 1 h and 24 h p.i. (Figure 5; Supplementary Figure 4).

Discussion
In this study, we explored the GEM-handle as a convenient method for designing combined nuclear and fluorescent tracers and evaluated its potential for the development of sdAb-based bimodal tracers.Due to the genetic encoding of the GEM-handle into the sequence of the targeting moiety, it is directly linked to the structure of the protein at a specific position and requires no separate synthesis.Additional advantages of the GEM-handle are the combination with simple site-specific labeling chemistry g u a r a n t e e i n g c o n s i s t e n t l a b e l p o s i t i o n i n g a n d t h e interchangeability of the fluorophore, which stands in contrast to trivalent platforms that necessitate comprehensive redesign and synthesis for fluorophore substitution.The labeling of sdAbs, which carried the GEM-handle, with IRDye800CW and subsequently 99m Tc, proved straightforward and successful.However, the amount of 99m Tc activity that could be used for radiolabeling had to be restricted to preserve the tracer's fluorescent signal.Hernandez et al. previously demonstrated that cyanine-based dyes are susceptible to radiobleaching and that this effect depends on the type of radiation and activity dosage (44).Although we observed an activity-dependent effect of 99m Tc on IRDye800CW, the extent of the effect was unexpected given the lower radiation energy of 99m Tc compared to isotopes such as 111 In, 68 Ga, or 121 Bi.To the best of our knowledge, this problem has not been described previously by other research groups combining cyanine-based fluorophores with 99m Tc, warranting further investigation (9)(10)(11)29).Hernandez et al. proposed the addition of scavengers, such as ascorbic acid, to prevent radiobleaching.However, these interfere with the redox reaction of the [ 99m Tc]Tctricarbonyl chemistry and adding scavengers post-radiolabeling proved futile as fluorescence was already compromised at that moment.We nevertheless demonstrated a linear up-scalability of the radiolabeling without effect on the fluorescence signal by increasing the tracer's mass and the reaction volume.A molar activity of 18.5 MBq/nmol enables in vivo studies in mice with sufficient tracer mass and radioactivity for fluorescent and radioactive tumor detection (12 nmol labeled with 222 MBq, for 6 mice).This molar activity is also clinically relevant as full-body SPECT/CT imaging and sentinel lymph node gamma-probing require respectively 370 -1110 MBq and 9.25 -18.5 MBq per patient (45-48), while 1 -10 mg (65-650 nmol) of fluorescent tracer is likely to be needed for intraoperative fluorescence detection.
The typical in vivo biodistribution of sdAb-based tracers is characterized by rapid renal clearance, leading to background signals mainly concentrated in the kidneys and bladder.This efficient clearance coupled with fast target recognition facilitates high-contrast imaging within 1 h p.i (27,34,36,(49)(50)(51)(52).The bimodal sdAb-GEM tracers presented in this study exhibit a comparable biodistribution profile except for a slightly elevated liver accumulation (approximately 2%ID/g).Furthermore, the targeted bimodal tracer only achieved sufficient tumor contrast 4 h p.i. instead of 1 h p.i.The hydrophobic nature of IRDye800CW, known for its binding to serum proteins and necrotic tissues, is thought to contribute to the non-specific liver accumulation (17,(53)(54)(55) specific liver accumulation, most likely due to the use of a more hydrophilic Cy5-based fluorophore (27).Towards the future, exploration of alternative near-infrared fluorescent dyes possessing improved in vivo pharmacokinetic behavior, enhanced radiostability, and a structure that does not interfere with the tricarbonyl chemistry could further enhance the potential of GEM-based bimodal tracers (52)(53)(54).
A limitation of this study was the relatively low tumor uptake values seen compared to the uptake values observed in the study conducted by Mateusiak et al., describing the generation and validation of the uPAR15 sdAb (40).This could possibly be explained by the use of a different tumor cell line (human colorectal HT-29 tumor cells versus human glioma U-87 cells).Consequently, we obtained smaller effect sizes, contributing to increased uncertainty in the statistical analysis.A direct comparison between uPAR15-GEM-IRDye800CW, uPAR15-GEM-[ 99m Tc]Tc(CO) 3 , and uPAR15-GEM-[ 99m Tc]Tc(CO) 3 -IRDye800CW in the same tumor model would be required to further assess the impact of the GEM-handle on the in vivo targeting of sdAb-based tracers.It remains however important to note that since this uPAR15 sdAb only recognizes the human homolog of uPAR, the murine uPAR expression on tumorassociated stromal cells in the tumor microenvironment does not contribute to the tracer's uptake, hereby underestimating the total tumor accumulation potential in a human situation (40,56).
The GEM-handle employed in this study consists of a hexahistidine-tag and a cysteine-tag separated by an amino acid spacer.The inclusion of a hexahistidine-tag in sdAbs (and other recombinant proteins) initially serves purification purposes (57, 58), however, it also offers the advantage of easy radiolabeling with 99m Tc through tricarbonyl chemistry (31).99m Tc proves to be an ideal radioisotope for the design of bimodal sdAb tracers given its wide availability via 99 Mo/ 99m Tc-generators, its half-life aligning well with the blood half-life of sdAbs (49), its relatively low radiation energy profile enhancing the safety for both patients and personnel, and its routine use for gamma-probing in the operating theatre, meaning all hardware, protocols and experience is available.Most often chelators, e.g.HYNIC and MAG3, are employed in clinic to prepare 99m Tc-labeled radiopharmaceuticals (59-63).Nevertheless, several compounds in clinical studies (64-66), including sdAbs (NCT 04483167, NCT 04040686) (67-69), make use of 99m Tc-tricarbonyl chemistry, showing the potential of this strategy for clinical translation.It is a relatively fast labeling procedure, a lyophilized kit is available, and it can be used in combination with any temperature-stable compound.
The inclusion of a cysteine-tag within the GEM-handle provides the ability for site-specific labeling using any maleimidefunctionalized fluorophore of interest.The widespread use of this chemical method ensures ready availability of such fluorophores.A drawback of the cysteine-tag is that it leads to a reduction in sdAb production yield and necessitates a reduction before fluorescence labeling (41).Enhancements to the described GEM-handle approach could involve the integration of alternative site-specific labeling motives, such as those based on enzymes or unnatural amino acids (70,71).
In conclusion, the GEM-handle is a convenient and fast method for designing and producing bimodal sdAb-based tracers, as well as any other tracer generated through fermentation or synthetic production.Further improvement of the GEM-design and conjugated fluorescent labels will increase its potential towards radiostability, good in vivo biodistribution and high contrast tumor imaging.