Abstract
1557
Objectives: Prostate-specific membrane antigen (PSMA) ligands such as 68Ga-PSMA have an established use in positron emission tomography (PET)/computed tomography (CT) imaging of patients with prostate cancer. However, the use of 18F-labeled PSMA ligands is increasing owing to their longer half-life, ease of larger batch production and lower positron range compared with 68Ga-labelled counterparts. Radiohybrid PSMA (rhPSMA) ligands are a new class of theranostic PSMA-targeting agents featuring fast 18F synthesis with utility for labelling with radiometals. Here, we evaluate the biodistribution and image quality obtained with the lead investigational compound, 18F-rhPSMA-7, to determine the optimum activity and imaging timepoint when imaging patients with prostate cancer.
Methods: Imaging data from clinically indicated 18F-rhPSMA-7 PET/CT scans captured in our institutions’ databases were retrospectively reviewed. Twelve categories were created based on the administered activity of 18F-rhPSMA-7 (low, 222-296 MBq; moderate, 297-370 MBq; and high, 371-444 MBq) and the uptake time (short, 50-70 min; intermediate, 71-90 min; long, 91-110 min, and longest, ≥ 111 min). SUVmean and organ/tumour-to-background ratio (ratio-SUVmean) were determined for background, healthy organs, and for 3 representative tumours. Overall image quality, non-specific blood pool activity and background uptake in bone/marrow were assessed using 3 or 4-point scales where low numbers indicated a favourable result. Kruskal-Wallis test (K-W) and ANOVA were used to estimate differences between groups. Linear-by-linear test, Kendal Tau-b, and Jonckheere-Terpstra (J-T) test were used to evaluate linearity between groups. Results: In total, 202 18F-rhPSMA-7 PET/CT scans from patients with prostate cancer (57 for primary staging and 140 for recurrent disease) Quantitative biodistribution in the liver, spleen, pancreas, and duodenum was similar across the range of uptake times and administered activities. Ratio-SUVmean decreased in the blood pool with increasing uptake time (1.20 ± 0.29; 1.21 ± 0.24; 1.22 ± 0.28; 1.10 ± 0.28, for short, intermediate, long, and longest, respectively; ANOVA p = 0.020, linearity p = 0.003) but increased in the kidney (48.09 ± 14.63; 55.80 ± 17.17; 55.98 ± 16.53; 58.95 ± 22.37, for short, intermediate, long, and longest, respectively; ANOVA p = 0.068, linearity p = 0.017), bladder (3.94 ± 2.16; 5.48 ± 5.22; 5.38 ± 2.79; 8.22 ± 5.71, for short, intermediate, long, and longest, respectively; K-W p < 0.001, J-T p < 0.001) and bone (1.90 ± 0.51; 2.12 ± 0.55; 2.17 ± 0.59; 2.34 ± 0.10, for short, intermediate, long, and longest, respectively; K-W p =0.027, J-T p = 0.001). In 377 tumour lesions, lesion-to-background ratios were not significantly different across all activity and uptake time groups. Qualitative image analysis revealed a statistically significant trend with increasing uptake time towards increasing non-specific bone/marrow uptake (2.23 ± 0.46; 2.33 ± 0.54; 2.378 ± 0.52; 2.65 ± 0.80, for short, intermediate, long, and longest, respectively; p = 0.011) and decreasing blood pool uptake (2.75 ± 0.51; 2.72 ± 0.61; 2.64 ± 0.67; 2.23 ± 0.82, for short, intermediate, long, and longest, respectively; p = 0.014). Overall image quality was highest in the low uptake time group; however, this was not statistically significant. Conclusions: Quantitative analysis shows uptake of 18F-rhPSMA-7 in tumour lesions is stable across multiple uptake times and administered activities. As low tracer retention in the bladder and bone marrow are important features for PSMA ligand imaging, and the highest overall image quality was achieved with short uptake times, the use of low activities (222-296 MBq) and an early imaging timepoint (50-70min) is recommended for 18F-rhPSMA-7.