The exact roles of PET in the imaging management of patients with known or suspected breast cancer are still in evolution. For assessing primary lesions, it is sometimes possible with PET to detect cancers occult on standard methods. This could be useful in high-risk patient populations, but in dense breasts, background FDG uptake is often higher than in women with fatty breasts, making identification of lesions < 1 cm in size improbable with current technologies. Distinguishing malignant from benign primary breast disease would seem better addressed by biopsy. With a positive predictive value of FDG PET for cancer over 96%, any FDG-avid breast lesion is highly suspicious and merits biopsy. Although PET in theory should be useful for depicting multifocal disease before surgery, the limitations in detecting small lesions in the breast limit the contribution of PET at present. It is most likely that PET will have a greater role in depicting primary breast lesions as dedicated PET imaging devices for the breast evolve. For axillary and internal mammary nodal staging, results with FDG PET are variable. Small nodal metastases < or = 5 mm will be missed by PET, whereas larger ones are more likely to be detected. PET can depict internal mammary nodes, but the accuracy of the method in this setting is not known, nor is there consensus on how identifying internal mammary node metastases will change treatment. Based on the available data, for pT1 breast lesions, PET, if negative, is not an adequate replacement for sentinel node surgery or axillary dissection. Results from the multicenter trial will be of great interest. Clearly PET can stage metastatic disease well. Bone scans with 18F- are exquisitely sensitive for metastases, and FDG is also very good. However, FDG PET can miss some blastic metastases to bone so at present FDG is not capable of excluding the presence of bone metastases. PET seems very well suited to detecting recurrences in soft tissues and the brachial plexus region in particular. The utility of PET in planning the treatment of individual patients appears promising. Although results must be confirmed in larger studies, it appears safe to conclude that failure of a chemotherapy regimen to decrease FDG uptake promptly in a breast cancer portends poor response. This does not hold true for hormonal therapy. At present, labeled estrogens are not widely available and cannot be recommended for clinical use. Thus, PET has shown considerable promise in breast cancer imaging, but in the author's experience is best applied to solve difficult imaging questions in specific patients and is not recommended for routine evaluation of the breast cancer patient. However, in larger primary tumors, the ability to use PET for staging and to plan treatment response suggest it will be more widely used. Additional studies with newer PET imaging devices and FDG and other tracers will help us better determine the role of PET in routine clinical care of the patient with known or suspected breast cancer. Certainly, this represent a fertile area for translational research studies over the next several years with the potential to significantly alter the way breast cancer is imaged and managed.