Impact of uptake time on image quality of [68 Ga]Ga-PSMA-11 PET/CT.

Abstract

With the introduction of prostate specific membrane antigen (PSMA) PET/CT, the detection rate of prostate cancer metastases has improved significantly, both for primary staging and for biochemical recurrence. EANM/SNMMI guidelines recommend a 60min time interval between [68 Ga]Ga-PSMA administration and acquisition. This study evaluates the possibility of a shorter time interval by investigating the dynamic change in image quality measures. We retrospectively analyzed 10 consecutive prostate cancer patients who underwent a dynamic whole body [68 Ga]Ga-PSMA-11 PET/CT of 75min from skull vertex to mid-thigh using Siemens FlowMotion. PET images were acquired directly after injection of 1.5MBq/kg [68 Ga]Ga-PSMA-11. Image quality measures included lesion maximum standardized uptake value corrected for lean body mass (SULmax ), tumor-to-background ratio (TBR), and contrast-to-noise ratio (CNR). Quantitative analysis of image quality in dynamic PET was performed using PMOD (version 4.2). Regions of interest (ROIs), drawn included different types of prostate lesions (primary tumor, lymph nodes, and bone metastasis), organ tissue (liver, spleen, lacrimal gland, submandibular gland, parotid gland, urinary bladder, kidneys blood pool [ascending aorta], left ventricle), bone tissue (4th lumbar vertebral body [L4]) and muscle tissue (gluteus maximus). To further investigate image quality four 10min multi-frame reconstructions with clinical parameters were made at different post-injection times (15, 30, 45, and 60min). A nuclear medicine physician performed a blinded lesion detectability evaluation on these multi-frame reconstructions for different prostate cancer lesions. Six primary prostate tumors in seven patients with prostate in situ, 13 lymph node metastases in six patients and up to 12 bone metastases in three patients were found. The different prostate lesion types (lymph nodes metastases, bone metastases, and primary prostate tumor) all show an increase in average SULmax , TBR, and CNR over time during the scan. The normalized average SULmax , TBR, and CNR of the combined prostate lesions at 15, 30, and 45min post-injection scans were all significant p<0.05 lower from the 60min post-injection [68 Ga]Ga-PSMA-11 PET/CT (9.5±4.5, 12.7±6.2, and 41.8±24.5, respectively). At patient level, the reader concluded the same regarding the presence/absence of primary prostate cancer recurrence, lymph node metastases, and/or bone metastases on all <60min post-injection [68 Ga]Ga-PSMA-11 PET/CT's in comparison to the reference scan (60min post-injection). At lesion level, all bone metastases seen on the reference scan were also seen on all <60min post-injection [68 Ga]Ga-PSMA-11 PET/CT's but there were some lymph nodes (n=2) metastases missed on the 15, 30, and 45min post-injection scans. One lymph node metastasis on both the 15 and 30min post-injection [68 Ga]Ga-PSMA-11 PET/CT's was missed and one lymph node metastasis was missed, only on the 45min post-injection [68 Ga]Ga-PSMA-11 PET/CT. Shorter post-injection times (15, 30, and 45min) compared to the recommended post-injection time of 60min are not optimal. However, the impact of a shorter time interval of 45min instead of 60min between [68 Ga]Ga-PSMA-11 administration and the start of PET/CT acquisition on both image quality (SULmax , TBR, and CNR) and lesion detection, while significant, is small.