Diagnostic Accuracy of Reduced Dose 18F-FDG PET/MRI in Imaging Assessment of Brown Adipose Tissue

1589 Objectives: 18F-FDG (FDG) PET/CT is currently the imaging method of choice for assessment of brown adipose tissue (BAT). Current PET/MRI with more sensitive solid-state PET detectors has the potential to significantly decrease radiation dose exposure and would be a more favorable BAT imaging modality, especially for younger and non-oncologic patients. The objective of this study was to determine if simulation of lower dose FDG administration with PET/MR could maintain the diagnostic accuracy for assessment of BAT. Methods: Healthy adult volunteers were prospectively enrolled on an IRB-approved clinical research protocol. Approximately, 2.59 MBq/kg (0.07 mCi/kg) ± 20% of FDG was administered intravenously, with minimum of 185 MBq (5 mCi) ± 20% and maximum of 555 MBq (15 mCi) ± 20%, which is about 50% of the standard dose for oncology patients. Prior to FDG injection, subjects were cooled in a temperature controlled room for one hour (15-16oC or 59-60oF) with a surgical cooling vest (14oC or 57oF) followed by continued cold exposure during the one hour uptake phase after FDG administration. PET/MR images were acquired from skull vertex through kidneys for a total of 5 minutes per field-of-view (100% imaging time) in a Signa 3T PET/MR system (GE Healthcare, Waukesha, WI) about 1.5 hours after FDG injection as per study protocol. Anonymized PET series were reconstructed from this raw PET/MR data for acquisition times reduced to 75% (3.75 minutes), 50% (2.5 minutes), and 25% (1.25 minutes), simulating FDG dose reduction. Qualitative and quantitative PET analysis was performed on Mirada XD PET software (Mirada Medical, Oxford, UK). BAT uptake was analyzed in each of the typical BAT anatomic distribution sites including bilateral cervical, supraclavicular/axillary, paraspinal, and mediastinal as well any miscellaneous regions. PET image quality was assessed on 5-point Likert scale. Qualitative BAT FDG uptake was visually graded according to a 5-point scale adapted from the Deauville scale used in lymphoma. Quantitative BAT FDG uptake was assessed for maximum (SUVmax) and peak (SUVpeak) standardized uptake value (SUV). Qualitative and quantitative BAT uptake at 100% imaging time were compared to each of the reduced acquisition times using Pearson’s correlation and paired T-test, respectively. Results: Study sample consisted of 9 subjects [56% male; mean age 27.3 years (range 22-36)]. All subjects had BMI less than 25 kg/m2 except one (35.9 kg/m2). A total of 54 anatomic regions with positive FDG BAT uptake were analyzed. PET image quality was good to excellent (Likert score 4 or 5 on a 5-point scale) for each analyzed BAT region, which did not vary across reduced imaging times. Qualitative visual BAT score was 4.54 ± 0.54 for the 100% imaging time and showed an excellent statistical correlation to each of the decreased acquisition times with all r > 0.9 and all p 0.05) given SUVpeak is less affected by image noise. Conclusion: Simulating FDG PET dose reduction using decreased FDG PET image acquisition times, we are able to show that FDG dose in BAT PET/MRI can be significantly reduced by up to 25% of our study research dose (and 12.5% dose of a standard oncologic FDG dose), reducing the whole-body effective dose equivalent to