In this review the findings of response monitoring studies in breast cancer, using [(18)F]2-fluoro-2-deoxy-D-glucose (FDG) and positron emission tomography (PET), are summarised. These studies indicate that there is a strong relationship between response and decrease in FDG signal even at an early stage of therapy. The review concentrates on methodological aspects of monitoring response with FDG: timing of serial scans, approach to region of interest definition, method of quantification and pitfalls of FDG. It is argued that, for clinical applications, there is now a need to standardise methodology. This would be necessary to establish firm cut-off values for discriminating responders from non-responders, which in turn would provide a means for providing optimal treatment for as many patients as possible.