Background The purpose of this study was to evaluate the imaging properties of hepatic metastases in 68 Ga-PSMA positron emission tomography (PET) in patients with prostate cancer (PC). Methods 68 Ga-PSMA-PET/CT scans of PC patients available in our database were evaluated retrospectively for liver metastases. Metastases were identified using 68 Ga-PSMA-PET, CT, MRI and follow-up scans. Different parameters including, maximum standardized uptake values (SUV max ) of the healthy liver and liver metastases were assessed by two- and three-dimensional regions of interest (2D/3D ROI). Results One hundred three liver metastases in 18 of 739 PC patients were identified. In total, 80 PSMA-positive (77.7%) and 23 PSMA-negative (22.3%) metastases were identified. The mean SUV max of PSMA-positive liver metastases was significantly higher than that of the normal liver tissue in both 2D and 3D ROI ( p ≤ 0.05). The mean SUV max of PSMA-positive metastases was 9.84 ± 4.94 in 2D ROI and 10.27 ± 5.28 in 3D ROI; the mean SUV max of PSMA-negative metastases was 3.25 ± 1.81 in 2D ROI and 3.40 ± 1.78 in 3D ROI, and significantly lower than that of the normal liver tissue ( p ≤ 0.05). A significant (p ≤ 0.05) correlation between SUV max in PSMA-positive liver metastases and both size (ρ Spearman = 0.57) of metastases and PSA serum level (ρ Spearman = 0.60) was found. Conclusions In 68 Ga-PSMA-PET, the majority of liver metastases highly overexpress PSMA and is therefore directly detectable. For the analysis of PET images, it has to be taken into account that also a significant portion of metastases can only be detected indirectly, as these metastases are PSMA-negative.

BackgroundThe purpose of this study was to investigate the imaging properties of pulmonary metastases and benign opacities in 68Ga-PSMA positron emission tomography (PET) in patients with prostate cancer (PC).Methods68Ga-PSMA-PET/CT scans of 739 PC patients available in our database were evaluated retrospectively for lung metastases and non-solid focal pulmonary opacities. Maximum standardized uptake values (SUVmax) were assessed by two- and three-dimensional regions of interest (2D/3D ROI). Additionally CT features of the lesions, such as location, morphology and size were identified.ResultsNinety-one pulmonary metastases and fourteen opacities were identified in 34 PC patients. In total, 66 PSMA-positive (72.5%) and 25 PSMA-negative (27.5%) metastases were identified. The mean SUVmax of pulmonary opacities was 2.2±0.7 in 2D ROI and 2.4±0.8 in 3D ROI. The mean SUVmax of PSMA-positive pulmonary metastases was 4.5±2.7 in 2D ROI and in 4.7±2.9 in 3D ROI; this was significantly higher than the SUVmax of pulmonary opacities in both 2D and 3D ROI (p<0.001). The mean SUVmax of PSMA-negative metastases was 1.0±0.5 in 2D ROI and 1.0±0.4 in 3D ROI, and significantly lower than that of the pulmonary opacities (p<0.001). A significant (p<0.05) weak linear correlation between size and 3D SUVmax in lung metastases (ρSpearman=0.207) was found.ConclusionBased on the SUVmax in 68Ga-PSMA-PET alone, it was not possible to differentiate between pulmonary metastases and pulmonary opacities. The majority of lung metastases highly overexpressed PSMA, while a relevant number of metastases were PSMA-negative. Pulmonary opacities demonstrated a moderate tracer uptake, significantly lower than PSMA-positive lung metastases, yet significantly higher than PSMA-negative metastases.