Purpose: The aim of this study was to assess the diagnostic performance of simultaneous whole-body Ga-DOTANOC PET/MRI compared with Ga-DOTANOC PET/CT for detection of distant metastatic disease in patients with well-differentiated neuroendocrine tumors (NETs).
Methods: Patients with histologically proven, well-differentiated NET (G1 or G2) were included in this prospective, institutional review board-approved study. Patients underwent Ga-DOTANOC PET/CT and subsequent Ga-DOTANOC PET/MRI after a single tracer injection on the same day for staging or restaging purposes. Images were evaluated for the presence of NET lesions by 2 rater teams, each consisting of a nuclear medicine physician and a radiologist, in an observer-blinded fashion. Overall agreement, accuracy, sensitivity, and specificity, relative to a composite reference standard (consensus review including follow-up data), were calculated.
Results: Between July 2014 and June 2016, 28 patients were enrolled. Overall agreement and accuracy between the 2 rater teams were 91.7% (95% confidence interval [CI], 87.5%-95.9%) and 97% (95% CI, 94.4%-99.6%) for PET/MRI and 92.3% (95% CI, 88.3%-96.3%) and 94.6% (95% CI, 91.2%-98.1%) for PET/CT, respectively (P = 1.00).Overall, PET/MRI reached 89.8% sensitivity (95% CI, 77.8%-96.6%) and 100% specificity (95% CI, 97%-100%); PET/CT showed 81.6% sensitivity (95% CI, 68%-91.2%) and 100% specificity (95% CI, 97%-100%) for the detection of metastatic disease in NETs.
Conclusions: Whole-body Ga-DOTANOC PET/MRI appears to be comparable to Ga-DOTANOC PET/CT for lesion detection in patients with well-differentiated NETs.