CMET-39. DISPARITIES IN STEREOTACTIC RADIOSURGERY USE AND OUTCOMES FOR MELANOMA BRAIN METASTASES – AN AUSTRALIAN POPULATION-BASED STUDY

Abstract

Abstract INTRODUCTION We aim to evaluate the use of stereotactic radiosurgery (SRS) among patients who received radiotherapy for melanoma brain metastases (MBM), and the associated outcomes. METHODS This is a population-based cohort of patients who received radiotherapy for MBM between 2013 and 2016, as captured in the Victorian Radiotherapy Minimum Dataset. Brain radiotherapy was classified as SRS (including multi-fraction stereotactic radiotherapy) and non-SRS. Mortality data was obtained through linkage with the Victorian Cancer Registry. The primary outcomes were: proportion of patients who had SRS for MBM, and overall survival (OS) following radiotherapy. Multivariate logistic regression was used to evaluate factors associated with SRS use, Kaplan Meier method for estimation of OS, and multivariate Cox regression for evaluation of factors associated with OS. RESULTS 294 patients received 551 courses of radiotherapy for MBM in this study, of which 39% (116/294) patients received SRS. Patients from higher socioeconomic status were more likely to have SRS – 49% in top quintile vs. 22% in lowest quintile (P< 0.009). Patients treated in regional centres were less likely to have SRS compared to metropolitan centres (3% vs. 48%,P< 0.001). In multivariate logistic regression, treatment in regional centres was the only factor independently associated with lower likelihood of receiving SRS (OR=0.04;95%CI=0.01–0.18;P< 0.001). The median follow-up of the cohort was 3.8 months (range: 0.1–42 months). There were 227 death observed, with 12-month OS of 23%. There was significant difference in 12-month OS between patients who received SRS compared to those who did not receive SRS (43% vs. 11%; P< 0.001). In multivariate Cox regression, the use of SRS was the only factor independently associated with OS (HR=0.38;95%CI=0.28–0.51;P< 0.001). CONCLUSION We observed sociodemographic and institutional disparities in SRS use for MBM. The observed impact of SRS use on OS is most likely confounded by patient selections for SRS.