Pmod v 4

Organ distribution of the radiotracer was determined in patient A by manually contouring the liver, gall bladder, kidneys, spleen, skeletal system, intestines, urinary bladder, and peripheral vein cannula (containing residual activity after tracer injection) in the CT for the attenuation correction using the PMOD software package (PMOD v4.0, PMOD Technologies LLC, Zurich, Switzerland). After subtracting remnant radiotracer activity in the peripheral venous catheter from the total prepared activity, the relative organ distribution (in %) could be calculated in relation to the determined organ volume in milliliter at 14 different time points (0.5, 1.5, 2.5, 3.5, 4.5, 5.5, 6.5, 7.5, 8.5, 9.5, 18, 30, 45, and 60 min after radiotracer injection). Mean and maximal standardized uptake values (SUV_mean and SUV_max) were determined by manually placing volumes of interest (VOIs) over representative areas in the liver and gall bladder at the above-mentioned 14 time points and in the kidneys, spleen, skeletal system, intestines, and urinary bladder at 18, 30, 45, and 60 min after radiotracer injection. Additionally, VOIs were placed in the HCC tumor lesions and non-tumorous liver tissue to determine SUV_max and SUV_mean. Measurements were performed on a syngo.via workstation (MM Oncology module, Siemens Healthineers, Erlangen, Germany).

For the quantitative analysis, dual-phase ¹⁸F-FBB (0–10 min, 90–110 min) and R1 PET-to-3D T1 MRI coregistration were performed initially for each subject separately using PMOD v4.0 (PMOD Technologies Ltd., Zurich, Switzerland). Voxel-wise parametric R1 maps were generated using PXMOD v4.0 with SRTM2 and a fixed subject-specific k2′ extracted from the regional kinetic modeling toolbox (PKIN) (Ottoy et al., 2019 (link)). SRTM2 was chosen over SRTM because the former model has been shown to reduce noise (Wu and Carson, 2002 (link)). Volumes of interest (VOIs) were delineated using an automated maximum probability atlas method (3 probability maps of gray matter, white matter, cerebrospinal fluid) for segmentation of each subject’s MRI and the Automated Anatomical Labeling (AAL) atlas in PMOD. Interframe motion correction was performed for early dynamic images. All VOIs included the cortical gray matter target regions (frontal, parietal, lateral temporal, anterior and posterior cingulate, and occipital cortices), and the reference region (cerebellum). Dividing the standardized uptake values (SUVs) of the different target regions by that of the reference region resulted in regional SUV ratios (SUVRs) for eFBB and dFBB. The composite value was defined as the arithmetic mean of the values of all target regions (Barthel et al., 2011 (link)).

Xenograft mice (n=5) were fastly injected of 200 µl of [¹⁸F]F-FGFR1 (11.1 MBq) through tail vein. For the in vivo blocking study, RT-112 (FGFR1-high expression) tumor-bearing mice (n = 5) were injected with a tracer that was mixed with 200-fold (1.58 mg) unlabeled peptide to saturate the receptors in the tumors. Mice were anesthetized via 3% isoflurane inhalation (RWD Life Science Inc. China) for induction and 2% isoflurane inhalation for maintenance. Micro-PET/CT imaging was performed on a micro-PET/SPECT/CT machine (NOVEL MEDICAL Equipment Ltd. China). Whole-body scans were performed 30, 60 and 120 min after tracer administration. The CT scanning parameters were 80 kV, 0.5 mA, 2000 ms/frames, and 360 frames; the images were reconstructed with a 512*512 matrix and a slice thickness of 0.18 mm. PET images were acquired with a 140*140 matrix and 10 min of total acquisition time. Reconstructed images were processed with the comprehensive image analysis software Pmod v4.201 (PMOD Technologies LLC. Switzerland). The biodistribution of the major organs (brain, heart, lung, liver, kidney, bone and muscle) and tumor were analyzed by manually outlining the regions of interest (ROIs) of the entire organs and tissue at each time point of the micro-PET/CT scan. The radioactivity accumulated in bone and muscle was measured in the limbs without xenograft tumors.