Study design for phase I

This study had an open-label design to assess the safety, whole-body biodistribution kinetics, and dosimetry of 99mTc-fucoidan in 10 healthy volunteers consisting of 5 male and 5 female subjects. 99mTc-fucoidan was synthesized according to GMP guidelines as described previously [14] and produced by Eurofins Amatsi Group (Fontenilles, France). The molar activity of 99mTc-fucoidan was 2994 ± 872 MBq/nmol at the end of synthesis, and the radiochemical purity exceeded 95% in every batch. Healthy subjects were intravenously injected with 356 ± 14 MBq of 99mTc-fucoidan, after undergoing baseline examinations. Whole body planar scintigraphy was performed at multiple time points after radiotracer administration (t = 30 min, 1.5 h, 3 h, 6 h and 24 h) and SPECT imaging at t = 2 h, based on previously reported blood clearance in a murine study [13]. Blood and urine analyses were performed before and after imaging.

The study was conducted in accordance with the Declaration of Helsinki and in compliance with current Good Clinical Practice guidelines. The protocols were approved by the local institutional review board, and all participants provided written informed consent. The study was registered as a clinical trial (NCT03422055, NL6703).

Whole-body scintigraphy

Planar whole-body scans were acquired with an energy peak at 140 keV, 15% energy window, table speed 8 cm/min, without automatic body contour and a matrix size of 256 × 1024 using low energy high resolution collimators. Dual-head SPECT was performed with the same energy peak, energy window and collimators as used for the whole-body scans with a matrix size of 128 × 128, 32 projections per detector; 40 s per angle, step mode and a noncircular rotation of the chest and the abdomen. All scans were performed on a Siemens Symbia-S.

Distribution kinetics and radiation dose estimates

All planar images were analyzed by drawing regions of interest for target organs using a dedicated software package (Olinda/EXM version 2.0, Hermes Medical Solutions, Stockholm, Sweden) [15, 16]. Whole-body scan derived values were used to determine biodistribution, residence times and dosimetry of 99mTc-fucoidan. Estimated organ doses were calculated from the geometric mean images, using the ICRP 89 Adult Male and Female phantom models in OLINDA/EXM. All biodistribution results are expressed as a percentage of the injected dose. A calibration source was used to convert the measured counts into the percentage injected dose activity. The same dose calibrator was used for the creation of the calibration source and for assay of injected subject doses. SPECT data were not used for biodistribution and dosimetry analysis. No attenuation correction was applied for these analyses.

The well counter used has a yearly calibration and validation of the most used radiotracers. In addition, we added for each subject an appropriate standard 99mTc activity to the measurements of the collected blood and urine samples obtained at different time intervals.

Regions of interest (ROIs) were drawn around the organs of interest to quantify tracer uptake [17]. The strategies we adopted to limit overlapping organs from distorting tracer uptake measurements can be found in the Supplemental Information. Whole-body region and bladder ROI were used to determine urinary excretion.

Safety assessment

Clinical assessment including vital signs and electrocardiography was performed before each imaging acquisition. Healthy volunteers were continuously monitored for adverse events. Blood samples were collected and analyzed. Analyses included hematology, hemostasis, and chemistry panels.

Blood and urine measurements

Urinary and venous blood samples were collected before and after 99mTc-fucoidan injection at set intervals. Urine samples were collected for 24 h, in the following intervals: 0–1, 1–3, 3–6 and 6–24 h after injection of 99mTc-fucoidan. Venous blood samples were collected in the following intervals: 0.5, 1.5, 3, 6, and 24 h after injection of 99mTc-fucoidan.

The radioactivity of whole blood was measured with a gamma counter (Wizard 2480, Perkin Elmer). Plasma was separated by centrifugation (3,000 g for 15 min at 4 °C), radioactivity was measured and decay-corrected for the time of injection. The total urine volume was measured, and a 1 mL sample was measured with a gamma counter. Radioactivity measurements were corrected for total volume of urine and duration of time intervals. The gamma counting measurements were done in three duplicate 1 mL samples of the collected blood and urine at different time intervals combined with an appropriate standard 99mTc activity to convert from counts to MBq.

Ex vivo binding study

To study 99mTc-fucoidan uptake in a disease-like setting, we incubated ex vivo human thrombus specimens with the tracer in the presence or absence of an excess amount of competitive binder. Venous blood samples were withdrawn from 11 healthy volunteers (6 males and 5 females, mean age 28.0 ± 1.8 years) into clot activator tubes (BD Vacutainer) and centrifuged for 15 min (3,810 g at 20 °C) to induce thrombus formation. Thereafter, serum was discarded and remaining thrombus was cut into cylinders of approximately 3 mm in thickness. Two cylinders per individual were used. As a positive control, the thrombus was incubated with 0.1 MBq of 99mTc-fucoidan (1.5 mL, tracer solution 0.07 MBq/mL) for 60 min at room temperature (RT). To assess the specificity of tracer binding, 1 µg (100 µL) of 3’-Sialyl-Lewis-X-tetrasaccharide (SLex, S1782, Sigma-Aldrich) was added two minutes before 99mTc-fucoidan in a parallel thrombus. The dose of SLex was calculated to be a 150-fold amount compared with the 99mTc-fucoidan. After incubation, the tracer solutions were discarded, and the thrombi were washed twice with 1.5 mL of phosphate-buffered saline (Fresenius Kabi, Netherlands). The radioactivity was measured using a gamma counter (2480 Wizard, Perkin Elmer) and radioactivity values were normalized for decay and weight of the thrombus. Results are depicted as counts per second/gram (CPS/g).

Deep vein thrombosis assessment

We studied whether 99mTc-fucoidan also binds to thrombi in vivo. This preliminary Phase IIa study was approved by the local institutional review board, and all participants provided written informed consent. The study was registered as NL7739. Thus, 3 patients with deep vein thrombosis (DVT), as diagnosed by routine diagnostic compression ultrasound sonography, were subjected to 99mTc-fucoidan SPECT imaging. All scans were performed within 48 h after treatment initiation. The patient cohort comprised individuals aged 50, 55, and 60 years, two of them being female.

Statistical analysis

Data are represented as means with standard deviations. To assess the difference in uptake for males and females an unpaired Student’s t-test was performed. For repeated measures data, we performed one-way ANOVA or mixed-model analysis. Statistical analyses were performed using the SPSS statistics software (Version 26, IBM). A p value of < 0.05 was considered statistically significant.