Background: To determine PET/CT and PET/MR reproducibility and PET/MR repeatability of fluorine 18 fluorodeoxyglucose (FDG) uptake measurements in tumors in cancer patients.
Methods: This IRB approved prospective study was performed between October 2015 and February 2016 in consecutive patients who performed same day PET/CT and two sequential PET/MR. Thirty three patients with visible tumors (N = 63) were included. SUV for body weight (SUV) and lean body mass (SUL) were obtained. Volume of interest (VOI) with a threshold of 40% was used and SUV/L's, metabolic tumor volume (MTV) and tumor to liver ratio (T/L) were calculated. Measurements were plotted in a scattered diagram to visually identify correlation, a regression line was drawn and the equation of the line was calculated. Bland-Altman plots expressed as percentages were constructed to assess the agreement between measurements. The maximal clinically acceptable limits range was defined as ±30%.
Results: Lesional SUV's, SUL's and MTV corrected to body weight (BW) and lean body mass (LBM) demonstrated strong positive linear correlation between PET/CT and PET/MR and between two sequential PET/MR. The 95% limits of agreement ranged from -27.7 to 17.5 with a mean of -5.1 and -27.6 to 17.9 with a mean of -4.9 for SUVpeak and SULpeak, respectively for sequential PET/MR. Other PET metrics demonstrated limits range that is above ±30% between PET/CT and PET/MR and between two sequential PET/MR.
Conclusion: PET/MR SUV/L peak has a clinically acceptable repeatability performance and can be used to evaluate the response to treatment.
Keywords: PET/MR; Reliability; Repeatability; Reproducibility; SUV.