The skeleton is the most frequent site of metastatic disease in breast cancer and also the site of greatest morbidity. In addition, there is now recognition that accelerated bone loss associated with chemotherapy or hormonal therapy leads to an increased risk of osteoporosis in long-term breast cancer survivors. An improved range of treatment options is available and assessment of skeletal response both to the disease and to therapy is therefore of growing importance. Plain radiographs remain widely used to assess response, but are of limited sensitivity. The isotope bone scan is more sensitive, but lacks specificity. Computerised tomography, magnetic resonance imaging and positron emission tomography all have an increasing role. In treatment-induced osteoporosis, bone mineral density is now readily measured by DEXA scanning. Tumour markers such as CEA, CA 15-3, CA 549 and TPA may have a role in assessing response, but probably in combination rather than individually, using an appropriate quantitative model. Several trials have shown that bone markers, especially markers of bone resorption such as Ntx, Ctx, PYD and DPD, appear to have strong potential as rapid, convenient and inexpensive measures of response. There is also evidence that they may be used as predictive or prognostic indicators. Evidence is accumulating that the reduction of bone resorption markers into the normal range results in substantially reduced morbidity in metastatic breast cancer and that this should be a major target of therapy.