Bone metastases often cause pain in cancer patients. Radiotherapy (RT) is the most effective treatment, but 20% of patients do not achieve adequate analgesia. The pathophysiology of malignant bone pain and mechanism for analgesia with RT are unclear. Our hypothesis is that cancer bone pain is due to altered bone turnover producing a chemical milieu that activates nociceptive fibres. RT relieves pain by affecting this process, and bone turnover marker levels can predict response to RT, reducing the burden of treatment for non-responders. Patients receiving localised RT for painful bone metastases are eligible for recruitment. Before RT and weekly for 4 weeks after, pain assessments, measurements of bone turnover markers Ntelopeptide (NTx), C-telopeptide (CTx), deoxypyridinoline (DpD) and bone alkaline phosphatase (BAP), and quantitative sensory testing (QST) are performed. To date, 115 patients have been assessed and 20 recruited. Results from 16 patients have been analysed. Overall, mean pre-RT values of all bone markers were raised. Differences were seen between responders and non-responders, particularly with CTx, NTx and BAP, although none was significant. No significant differences were found with QST. We feel initial results suggest support for our hypothesis, but the lack of significance so far is partly due to the sample size. Recruitment continues. In view of the small number of patients recruited, we have also analysed patient demographics and reasons for non-inclusion or incomplete data. We will discuss these results and consider them in modifying existing studies and future research. We must obtain scientific evidence to support best practice and development of palliative medicine; balancing this with caring for an often fragile population is an important challenge.