Abstract
RATIONALE AND OBJECTIVES: Prone (18)F fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT) may have advantages for breast imaging because of improved separation of deep anatomic structures. There are limited data on whether prone and supine FDG-PET/CT provide similar information regarding breast and axillary disease in the setting of locally advanced breast cancer (LABC). The purpose of this study was to compare the information on locoregional disease distribution provided by prone versus supine FDG-PET in newly diagnosed LABC. MATERIALS AND METHODS: In an Institutional Review Board-approved prospective trial, 24 patients with newly diagnosed LABC underwent both supine and prone FDG-PET/CT at the same scanning session. Three readers performed an independent review of all scans and categorized the locoregional disease distribution as breast only (BO)-unifocal, BO-multifocal, BO-multicentric, or breast + axillary involvement. For breast + axillary disease, the readers also assessed the number of involved axillary lymph nodes. Interobserver discrepancies were resolved at a consensus reading session. RESULTS: Two scanning sessions were excluded because the prone scan had omitted part of the axilla from the field of view. In the remaining 22 patients, the consensus categorization of anatomic disease distribution was concordant between prone and supine scanning in 21 patients (linear kappa 0.91, 95% confidence interval [0.79-1]). In the 16 patients with breast + axillary disease, equal numbers of involved lymph nodes were identified on prone and supine scanning in 12 patients, whereas in the remaining four patients, prone scanning resulted in a higher number of visualized lymph nodes. CONCLUSIONS: Prone and supine FDG-PET/CT provided statistically identical information on locoregional disease distribution in LABC. However, prone scanning may perform better than supine for assessing the number of involved lymph nodes. Prone FDG-PET/CT may be useful in future clinical and research efforts, including hybrid PET-magnetic resonance imaging (MRI) applications.