Classification of abnormal findings on ring-type dedicated breast PET for the detection of breast cancer

Abstract

Aim: To investigate the usefulness of classification of ring-type dedicated breast positron-emission tomography (dbPET) findings in detection of breast cancer. Patients and Methods: A total of 709 patients with breast cancer underwent dbPET before treatment. Each finding was morphologically categorized as a focus (uptake size ≤5 mm), mass (>5 mm), or non-mass (multiple uptakes not belonging to a three-dimensional mass or without distinct mass features). Non-mass uptakes were additionally classified as linear, focal, segmental, regional, or diffuse distributions. Lesion-to-background ratios were calculated. Results: Among 910 abnormal findings, 700 (76.9%) were malignant and 210 (23.1%) were benign. Morphologically, 198 (21.8%) lesions were foci, 431 (47.4%) were masses, and 281 (30.9%) were non-masses. In multivariate analysis, mass, focal and segmental distributions of nonmass lesions and high lesion-to-background ratio were significantly related to breast cancer (all p<0.001). Conclusion: Classification of abnormal findings on dbPET using morphology and lesion-to-background ratio were useful to detect breast cancer.