Many patients with incurable cancer will receive palliative oncological treatment before their death, and radiotherapy (RT) is an important element of this. The aim of palliative RT is to alleviate symptoms and improve quality of life. An accurate and practical survival prediction model for metastatic cancer patient receiving palliative RT can assist the decision making (ranging from best supportive treatment alone for expected short survival, to dose escalation for potential better disease control).

The available survival prediction models (such Survival Prediction Score using Number of Risk Factors by Chow et al and TEACHH model) have been developed in the Western world. We therefore perform a prospective observational study 1) to assess the overall survival of patients evaluated for palliative RT at a tertiary hospital in Hong Kong, and 2) to validate the prognostic score systems in our population.

Inclusion Criteria:

• All adult patients with stage IV cancer who are referred for palliative RT in Tuen Mun Hospital, including both inpatients and outpatients.

Exclusion Criteria:

• Patients who have received palliative RT before (i.e. not the first course palliative RT), have non-metastatic disease, are misclassified as palliative patients.

Stage IV cancer patients evaluated for palliative radiotherapy in Tuen Mun Hospital

Demographic data, and the information required for the prognostic scores will be extracted (e.g. presence of liver metastasis). Also data on the date of birth and death, sex, primary cancer. Data on the primary site of disease were coded using International Classification of Disease version 9.

Personal identifier will be deleted to protect patient confidentiality.

After publication.