*4.4. Al18F-Labelling of the Nb-RESCA*

100 μL (1 GBq) of [18F]NaF solution was added to 10 μL (20 nmol) of AlCl3 solution (2 mM AlCl3 trace-metal in 0.1 M NaOAc buffer pH 4.1) and left at RT for 5 min. The RESCA-conjugated cAbVCAM1-5 Nb (43 nmol, stock solution of 86.2 μM) in 0.1 M NH4OAc pH 4.5 was added to the reactor containing the previously prepared [18F]AlF solution and left at RT for 12 min.

### *4.5. Purification and Quality Control*

The radiolabelled Nb was purified using a disposable PD-10 column (GE Healthcare, Machelen, Belgium) pre-conditioned with injection buffer (0.9% NaCl + 5 mg/mL vit. C, pH 6), and passed through a 0.22 μm PVDF membrane filter (Millex GV, Millipore, Darmstadt, Germany) before further use. RCP was assessed through instant thin layer chromatography (iTLC) on silica gel impregnated glass fiber sheets (Agilent Technologies, Machelen, Belgium) with 0.9% NaCl as mobile phase. iTLCs were analysed with a radio-TLC detector (RITA, Elysia Raytest, Angleur Belgium). RCP before in vivo injection was >99%. DC-RCY was calculated based on the activity obtained after PD-10 to the amount of starting activity used for Al18F-production, decay-corrected for the same time point.

### *4.6. In Vitro Stability Studies*

At different time points, aliquots of filtered [18F]AlF(RESCA)-cAbVCAM1-5 Nb were analysed for stability in injection buffer or in serum. For the latter, 400 μL of filtered [18F]AlF(RESCA)-cAbVCAM1-5 Nb in injection buffer was added to 500 μL of human serum (Innovative research, Peary Court, FL, USA) and incubated at 37 ◦C. Analyses were performed via SEC on a HPLC system (Hitachi, Zaventem, Belgium) equipped with a radio-detector (GABI, Elysia Raytest, Angleur, Belgium) and on a Superdex

75 10/300 GL column (GE Healthcare, Machelen, Belgium) equilibrated with Phosphate Buffer Saline (PBS) ([18F]AlF(RESCA)-cAbVCAM1-5 Rt = 28.6 min, free [18F]AlF and [18F]F-Rt = 35.4 min).

#### *4.7. Animal Model and Experimental Setup*

All animal experiments were performed in accordance to the European guidelines for animal experimentation under the license LA1230272 and approved by the local Ethical Committee of the Vrije Universiteit Brussel (14-272-7). ApoE−/<sup>−</sup> mice were obtained from Charles River (L'Abresle, France). ApoE−/<sup>−</sup> mice were fed a high-fat Western diet with 1.25% cholesterol (D12108C, Research Diets, New Brunswick, NJ, USA) for 25–30 weeks to induce atherosclerotic lesions. VCAM-1 expression was assessed for this model in a previous study [14]. ApoE−/<sup>−</sup> mice (*N* = 6) were injected IV with (14.52 <sup>±</sup> 8.98) MBq (12 <sup>μ</sup>g) of [18F]AlF(RESCA)-cAbVCAM1-5 Nb. The control group (*N* = 6) was injected with an excess of unlabelled cAbVCAM1-5 Nb (1.1 mg, 90-fold excess) followed by the injection of (14.34 <sup>±</sup> 8.28) MBq (12 <sup>μ</sup>g) of [18F]AlF-RESCA-cAbVCAM1-5 Nb 15 min after the first injection.

#### *4.8. In Vivo PET*/*CT Imaging and Image Processing*

Two hours and 30 min after tracer injection (as determined to be the optimal time point for ideal T/B ratio in a previous study) [14], mice (*N* = 6/group) were imaged sequentially on two different PET/CT systems using a cross-over design: a β-CUBE (Molecules, Ghent, Belgium) providing sub-mm (0.83 mm) spatial resolution and a LabPET8 (TriFoil Imaging, Chatsworth, CA, USA) with 1.2 mm spatial resolution. Both PET scans were acquired in list-mode with a total acquisition time of 30 min on the β-CUBE and 30 min on the LabPET8. PET data were reconstructed iteratively (OSEM for β-CUBE; MLEM for LabPET8) with a total of 50 iterations into a voxel size of 0.4 and 0.5 mm for the β-CUBE and LabPET8, respectively. Each PET scan was followed by a CT scan acquired for co-registration purposes using the CT-device of the same manufacturer (X-CUBE for β-CUBE; XO-CT for LabPET8). Volumes-of-interest were drawn at the level of the aortic arch, brain and heart, and T/B and T/H ratios were calculated.

#### *4.9. Ex Vivo Analysis*

Following in vivo PET/CT imaging, mice were euthanised to collect organs and tissues of interest. Aortas from the aortic root to the iliac bifurcation were collected as well. All samples were weighed and counted for radioactivity against a standard of known activity. Uptake was expressed as percentage of injected activity per gram (%IA/g), corrected for decay and extra-venous injection. Ex vivo autoradiography images were obtained after overnight exposure of the aortas to a dedicated phosphorscreen (Typhoon FLA 7000, GE Healthcare). Images were analyzed with ImageQuant.

#### *4.10. Data Analysis and Statistics*

Data are expressed as mean ± standard deviation. Comparisons between groups were performed using unpaired Student's t-test; for comparisons between scanners a paired Student's t-test was used. A *p*-value ≤ 0.05 was considered significant. Statistical analysis was performed using Prism 5 (Graph Pad Software) or SPSS Statistics software (version 24.0.0, IBM Company, Brussels, Belgium).

#### **5. Conclusions**

The cAbVCAM1-5 Nb could be easily radiolabelled with [18F]AlF through chelation with RESCA. The potential of 18F-labelled cAbVCAM1-5 Nb to target the atherosclerotic lesions and to provide good target-to-background ratios was demonstrated using the new β-CUBE imaging system. However, in vivo degradation leads to bone uptake of fluorine-18, which could interfere with the interpretation of imaging results. The excellent and uniform spatial resolution of the β-CUBE resulted in improved image quality and allowed better quantification as compared with an imaging system with lower resolution.

**Supplementary Materials:** The following are available online, Figure S1: Mass determination analysis of the modified Nb, Figure S2: SEC profile of [18F]AlF(RESCA)-cAbVCAM1-5 Nb in human serum, Table S1: Biodistribution.

**Author Contributions:** Writing—original draft preparation, J.B.; writing—review and editing, S.H., C.V.; image acquisition & system handling, B.D; Image analysis, J.B., B.D., S.N., C.V., S.H.; Nanobody development, N.D., A.B.; radiochemistry, J.B., F.C.; data acquisition, J.B., P.D.; supervision, G.B., V.C., C.X., N.D., S.H.; All authors have read and agreed to the published version of the manuscript.

**Funding:** This project has received funding from FWO project N◦G005815N and G0D8817N. This work was funded by a grant from the Scientific Fund W. Gepts UZ Brussel. J. Bridoux is funded by the EU H2020 MSCA ITN PET3D. Frederik Cleeren is a Postdoctoral Fellow of FWO (12R3119N).

**Acknowledgments:** The authors thank Cindy Peleman for technical assistance.

**Conflicts of Interest:** S. Neyt is an employee of MOLECUBES NV. A. Broisat and N. Devoogdt have patent on VCAM Nanobodies (PCT/EP2012/066348), granted amongst other countries in US and Europe (US9771423B2 and EP2748196B8)
