Simultaneous whole body 18F-fluorodeoxyglucose positron emission tomography magnetic resonance imaging for evaluation of pediatric cancer: Preliminary experience and comparison with 18F-fluorodeoxyglucose positron emission tomography computed tomogra

AIM: To describe our preliminary experience with simultaneous whole body 18F-fluorodeoxyglucose(18F-FDG)positron emission tomography and magnetic resonance imaging（PET-MRI） in the evaluation of pediatric oncology patients.METHODS: This prospective, observational, singlecenter study was Health Insurance Portability and Accountability Act-compliant, and institutional review board approved. To be eligible, a patient was required to:（1） have a known or suspected cancer diagnosis;（2） be under the care of a pediatric hematologist/oncologist; and（3） be scheduled for clinically indicated 18F-FDG PETCT examination at our institution. Patients underwent PET-CT followed by PET-MRI on the same day. PET-CT examinations were performed using standard department protocols. PET-MRI studies were acquired with an integrated 3 Tesla PET-MRI scanner using whole body T1 Dixon, T2 HASTE, EPI diffusion-weighted imaging（DWI） and STIR sequences. No additional radiotracer was given for the PET-MRI examination. Both PET-CT and PETMRI examinations were reviewed by consensus by two study personnel. Test performance characteristics of PETMRI, for the detection of malignant lesions, including FDG maximum standardized uptake value（SUVmax） and minimum apparent diffusion coefficient（ADCmin）, were calculated on a per lesion basis using PET-CT as a reference standard.RESULTS: A total of 10 whole body PET-MRI exams were performed in 7 pediatric oncology patients. The mean patient age was 16.1 years（range 12-19 years） including 6 males and 1 female. A total of 20 malignant and 21 benign lesions were identified on PET-CT. PET-MRI SUVmax had excellent correlation with PET-CT SUVmax for both benign and malignant lesions（R = 0.93）. PETMRI SUVmax > 2.5 had 100% accuracy for discriminating benign from malignant lesions using PET-computed tomography（CT） reference. Whole body DWI was also evaluated: the mean ADCmin of malignant lesions（780.2 + 326.6） was significantly lower than that of benign lesions（1246.2 + 417.3; P = 0.0003; Student’s t test）. A range of ADCmin thresholds for malignancy were evaluated, from 0.5-1.5 × 10-3 mm2/s. The 1.0 × 10-3 ADCmin threshold performed best compared with PETCT reference（68.3% accuracy）. However, the accuracy of PET-MRI SUVmax was significantly better than ADCmin for detecting malignant lesions compared with PET-CT reference（P < 0.0001; two-tailed Mc Nemar’s test）.CONCLUSION: These results suggest a clinical role for simultaneous whole body PET-MRI in evaluat