Objective Patients with non-curative malignancy can receive palliative radiotherapy (PR) to alleviate symptoms. However, choosing the right patient to receive PR can be challenging, as some patients may not survive long enough to gain benefit. This study aims to identify prognostic factors for overall survival (OS) and 30-day mortality (30DM) following PR and to test these in a real-world cohort.

Method A retrospectively collected data set of all adults completing PR between 1 August 2018 and 31 December 2018 at a single centre (n=214, Southend University Hospital NHS Foundation Trust, UK) was used to test prognostic factors. Factors such as demographics, tumour primary, treatment area, fractionation regime, performance status (PS), progressive disease (PD), opioid or steroid use and haemoglobin level, as well as overall survival, were collected. Cox regression was used to examine survival predictors, and logistic regression was used to determine the predictive strength of factors for 30DM.

Results Overall 30DM was 14%. There was significantly worse survival in patients with poor PS (HR 1.2406, 95% CI 0.94 to 1.64. p=0.01). Patients with PS 3 had a median OS of 75 days and were more likely to experience 30DM (OR 6.2, 95% CI 1.226 to 45.42, p=0.03). Patients with PD outside of the radiation field (46%, 30 out of 65 documented) had significantly worse OS (HR 5.24, 95% CI 2.19 to 12.5, p<0.001).

Conclusion Poor PS and PD were prognostic of OS and 30DM. Future work should include validation with a prospectively collected cohort.