*4.2. Imaging Protocol and Analysis*

68Ga-labeled PSMA ligand, Glu-urea-Lys(Ahx)-HBED-CC ([68Ga]Ga-PSMA-11), was synthesized using sterile methods as previously described by Eder et al. [36]. The included patients underwent imaging on a Biograph 64 TruePoint (True V HD) PET/CT scanner (Siemens/Erlangen/Germany) 60 ± 10 min (whole body; baseline scan) p.i. of 195.5 ± 48.3 MBq (median activity: 193 MBq, range: 97–299 MBq) and 180 ± 10 min. p.i. (pelvic, abdominal, and suspicious regions; delayed scan). The following parameters were used: three-dimensional acquisition mode (168 × 168); acquisition time of three min. per bed position; axial field of view (FOV): 21.8 cm; random, scatter, and decay correction; ordered-subsets expectation maximization method (OSEM) for PET image reconstruction (two iterations, 14 subsets, Gaussian filtering, 4.2 mm transaxial resolution, full-width at half-maximum). Attenuation corrections were performed using the low-dose non-enhanced CT data (120 kV, 20–60 mAs, CT transverse scan-field 50 cm, 70 cm extended FOV, resolution 1.0 s, 0.6 mm) or the contrast-enhanced CT data (140 kV, 100–400 mAs, dose modulation). The images were assessed by nuclear medicine clinicians and radiologists (each with more than 5 years experience in PET/CT imaging) and reviewed visually in consensus by two board-certified nuclear medicine clinicians and one board-certified radiologist. The term "lesion positivity rate" is used based on the imaging result and its interpretation by the nuclear medicine and radiological expert team in relation to the PSMA-positive tumor lesions. Any lesion with an increased radiotracer uptake (measured with SUVmax) above physiological uptake was considered suspicious of malignancy, and biphasic lesion detection (baseline and delayed images) was taken into account. If no consensus could be found between the board-certified nuclear medicine clinicians and the board-certified radiologist, these lesions were classified as moderately suspicious of malignancy. However, in this case, all the experts classified the lesions as abnormal and probably malignant. SUVmax of PSMA-avid lesions detected by baseline and by delayed scan were compared. There have been extensive efforts to develop quantitative criteria for the analysis of oncologic PET images. The measures proposed are based on the SUV in a certain volume of interest (VOI) enclosing the lesion. The principle of the SUV was introduced by Strauss and Conti [37]. For a defined VOI, the mean SUV value (SUVmean) of all included pixels is usually calculated as a representative measure of tracer uptake. As a result of the VOI definition dependence, the SUVmean suffers from a limited reproducibility. To overcome this problem, the SUVmax has been introduced, which is the maximal SUV value in the lesion. Thus, we have only used the SUVmax in the present study. LNM were divided into two groups based on their location (pelvic LNM: iliac and/or pararectal) and extra-pelvic distant LNM (retroperitoneal and/or above the iliac bifurcation).
