Sinograms were framed into 25 frames (6 × 10 seconds, 4 × 15 seconds, 2 × 30 seconds, 2 × 120 seconds, 1 × 180 seconds, 6 × 300 seconds, and 4 × 600 seconds) and reconstructed with an ordered subset expectation maximation (OSEM) two-dimensional iterative algorithm. Images were summed from 60 to 80 minutes, and volumes of interest were drawn over whole tumors with Inveon Research Workplace image analysis software 4.1 (Siemens Medical Solutions), using the CT template as an anatomic reference. Radioactivity uptake was calculated as the percentage of injected dose per gram tissue (%ID/g) in whole tumor. After PET/CT scans, mice were killed, tumors were collected,

and paraffin-embedded tumor sections were stained with antibodies raised against CA IX (ab15086; Abcam, Cambridge, UK, 1:8000), Glut-1

(GT12-A; Immune Diagnostics and Research, Hämeenlinna, Finland, 1:1000), and Hif-1α (610959; BD Transduction Laboratories, Franklin find more Lakes, NJ, USA, 1:100). Immunostaining was performed as described previously [11]. The expressions of CA IX, Glut-1, and Hif-1α were visually analyzed from 10 different areas in each tumor section using a × 20 microscope objective. The percentage of positively stained tumor cells was counted, and staining intensity was described as weak (1), moderate (2), or strong (3). Each tumor was scored (range = 0-300) by multiplying the average intensity value by the average percentage of positively stained cells. Analyses were performed independently by two investigators (J.S. and K.K.). Digital autoradiography

was used to measure the EX 527 nmr uptake of [18F]EF5 and [18F]FDG in cell lines grown on eight-well chamber slides (Nunc™, Thermo Scientific, Waltham, MA, USA) under normoxia and different time periods of hypoxia (1% O2). Four days before tracer incubation, cells were plated onto chamber slides in duplicate (two wells per cell line) and cultured under normal culturing conditions. Medium Nintedanib (BIBF 1120) was changed on the third day. Cells were seeded at various densities according to their growth rates to ensure that there would be equal amount of cells at the time of tracer incubation. On day 4, one chamber slide per time point was transferred to a hypoxia workstation (Invivo2; Ruskinn Technology Ltd, Pencoed, United Kingdom) at 24, 12, 6, 3, and 1 hour before the end of tracer incubation time. Chambers were removed, and slides were washed with phosphate-buffered saline before being incubated with [18F]EF5 or [18F]FDG for 60 or 30 minutes, respectively, in 50 kBq/ml Dulbecco’s modified Eagle’s medium ([18F]EF5) or physiological saline ([18F]FDG) under hypoxia. The workstation and all solutions used were stabilized in 1% O2 before the experiment. Control cells were incubated with tracers under normal culture conditions (normoxia, 0 hour). Cells were then washed with phosphate-buffered saline and fixed in 4% paraformaldehyde for 10 minutes at room temperature.