Ultra-low-dose 18F-FDG PET/CT protocols, quality measurements and experience in clinical patient management

Objectives Increasing use of 18F-FDG for diagnosis, staging and therapy monitoring of cancer has led to concern on total injected radiation dose. We investigated the impact of reduced dose on quantitative accuracy, and in a clinical study, using time of flight PET/CT system.

Methods Phantom measurements (i) and clinical patient evaluation (ii) were performed. In the first, a NEMA body phantom was scanned four times, with total activity (TA) of 20 and 40 MBq, and target-to-background ratio (TBR) of 4:1 and 2:1. Five lower activity images were simulated by truncating list-mode files (1-3min acq. times), using default recon parameters. ROIs were defined on CT, and observed activity statistics recorded. In the second study, all patients (n=80) imaged using the low-dose protocol (<=2.7 MBq/kg, 2.4±0.3) between Sept 2008 and May 2011 were enrolled retrospectively.

Results All targets were visible on all phantom images, except the 10mm sphere at 2:1 TBR. Excluding this one, no significant differences of observed contrast were seen between 20 and 40 MBq TA for either TBR. In the patient study, 31 scans were PET-positive, and 49 PET-negative. Histopathological confirmation (20 pts) and clinical follow-up data (all pts,17.3±7 months) were obtained, and 25 had one or more malignant lesions (positive), 55 were negative. PET provided 22 true positive, 46 true negative, 9 false positive (due to inflammation or benign disease e.g. adrenal adenoma) and 3 false negative cases (2 low activity tumours, 1 intestinal activity), consistent with confounding effects as reported at higher doses.

Conclusions Considering the TOF capabilities of new PET/CT scanners, a clinical protocol was implemented with mean injected activity of 2.7 MBq/kg, 1.5 mins/bed-position. The accuracy of lesion detection was similar to what is reported in literature using conventional PET with higher doses