ESTRO 2025 - Abstract Book

S690

Clinical - CNS

ESTRO 2025

Purpose/Objective: The number of people living with brain metastases (BMs) has increased in the last decade[1], driven by availability of high-quality MRI for diagnosis and improved survival from extracranial and intracranial disease. SRS has replaced WBRT for many patients, with excellent local control and fewer side-effects[2]. Despite the large number of patients treated with SRS, there remains significant variability in the radiation dose prescribed[3]. There are very few prospective studies of the dose-response relationship to guide practice[4], and no data from randomised control trials[3]. As SRS becomes more widely used, it is important to optimise technical delivery to improve patient outcomes. We present preliminary dose-response data from a prospective observational study of post-SRS outcomes in patients with BMs (SAFER, IRAS 276567). Material/Methods: The first 100 patients recruited to SAFER were included. They were treated using single and fractionated SRS and had 3-monthly surveillance MRIs thereafter. Nine months of outcome data were included in this planned preliminary analysis. Local control probability was calculated using Kaplan Meier methodology. Prescription doses were converted to Biologically Effective Doses (BED) using the linear quadratic equation, and an alpha-beta ratio of 10. Logistic regression was used for dose-response modelling on a per-lesion basis. Statistical analysis was performed on Prism v10.2.3. Results: 499 BMs were included (Table 1). The overall 9-month local control rate (LCR) was 91.59% (Figure 1a). In those who received BED 10 >60, the LCR was 93.95%, compared to 88.25% for those who received BED 10 ≤60, although this difference was not statistically significant (p=0.0744). The dose-response curve demonstrated a significant relationship between BED 10 and LCR (odds ratio 1.072, slope 3.237, p=0.0012) (Figure 1b), and acceptable model discrimination (AUC 0.6790, p=0.0069) (Figure 1c). Based on the regression model, 72.8% of those receiving BED 10 40 were predicted to have local control at 9 months. This increased to 97.7% in those receiving BED 10 80. The dose response relationship remained significant even after tumour size was incorporated into multivariable analysis (odds ratio 1.075, slope 2.784, p=0.0054) (results not shown).