The role of stereotactic radiotherapy in brain metastases from gastrointestinal cancer.

181 Background: The incidence of brain metastases from Gastrointestinal primary tumors (GI-BMs) is approximately of 6%. Appropriate management of these patients, who often presents at advanced disease stage, is poorly defined. Stereotactic Radiotherapy (SRT) could be proposed to improve symptoms and extend disease control. The aim of this study is to assess the impact of SRT in this population. Methods: Data from consecutive patients treated for GI-BMs with SRT from March 2015 to March 2024 were retrospectively collected. Dose was expressed as Equivalent Dose in 2Gy Fractions (EQD2). lntra-Cranial Control (IC) and Overall Survival (OS) were calculated from date of SRT to event (intracranial relapse and death respectively) or last follow-up. Results: Fifty-four patients (median age 68 years, range 46-84) accounting for 98 GI-BMs were included. Primary tumors were colorectal and gastro-oesophageal adenocarcinoma in 37 (68.5%) and 17 (31.5%) patients respectively. Eighteen (33%) patients presented with a single GI-BM. Extra-cranial disease was present in 41 (76%) patients, including unresected/relapsing primary tumor in 9 (17%) cases. Detection of BMs occurred at diagnosis in 8 (15%) patients, as first site of relapse in 27 (50%) or as progression of known metastatic disease in 19 (35%) patients. Before SRT, surgery and/or WBRT were performed in 10 (19%) and 5 (9%) patients respectively. For each SRT course a median of 1 (range 1-5) GI-BMs was treated. Dose regimens consisted of 12-30 Gy in 1-5 fractions, to a median EQD2 of 50 Gy 10 (range 22-68). A second SRT course for out-of-field intracranial progression was performed in 7 patients (13%). First- and further chemotherapy lines were administered in 39 (72%) and 15 (28%) patients respectively. IC rate was 88% at 6 months and 70% at 1 year. At univariate analysis (UVA) no variables were correlated with IC. Median OS was 11 months (IC 95% 9-16), 6-months and 1-year OS rate was respectively 76% and 48%. Only systemic treatment beyond first line was correlated with OS at UVA (4 versus 15 months, p=0.02). Two (4%) patients developed mild symptomatic radionecrosis. Conclusions: Onset of GI-BMs can occur at any stage of disease. Despite poor overall outcome, selected patients may benefit from SRT to improve IC, particularly in association with first-line systemic treatment. Multiple SRT courses can be delivered in case of further intracranial recurrence. Symptomatic radionecrosis may occur in less than 5% of cases.